5.4.106-rc1 review
Subsystems (3)
[email protected]
maintainercover 5.4.106-rc1 review
From: Greg Kroah-Hartman <[email protected]>
This is the start of the stable review cycle for the 5.4.106 release. There are 168 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 Wed, 17 Mar 2021 13:55:26 +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-5.4.106-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-5.4.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 5.4.106-rc1
Juergen Gross <[email protected]>
xen/events: avoid handling the same event on two cpus at the same time
Juergen Gross <[email protected]>
xen/events: don't unmask an event channel when an eoi is pending
Juergen Gross <[email protected]>
xen/events: reset affinity of 2-level event when tearing it down
Marc Zyngier <[email protected]>
KVM: arm64: Reject VM creation when the default IPA size is unsupported
Marc Zyngier <[email protected]>
KVM: arm64: Ensure I-cache isolation between vcpus of a same VM
Keith Busch <[email protected]>
nvme: release namespace head reference on error
Keith Busch <[email protected]>
nvme: unlink head after removing last namespace
Marc Zyngier <[email protected]>
KVM: arm64: Fix exclusive limit for IPA size
Josh Poimboeuf <[email protected]>
x86/unwind/orc: Disable KASAN checking in the ORC unwinder, part 2
Lior Ribak <[email protected]>
binfmt_misc: fix possible deadlock in bm_register_write
Naveen N. Rao <[email protected]>
powerpc/64s: Fix instruction encoding for lis in ppc_function_entry()
Mathieu Desnoyers <[email protected]>
sched/membarrier: fix missing local execution of ipi_sync_rq_state()
Minchan Kim <[email protected]>
zram: fix return value on writeback_store
Alexey Dobriyan <[email protected]>
prctl: fix PR_SET_MM_AUXV kernel stack leak
Matthew Wilcox (Oracle) <[email protected]>
include/linux/sched/mm.h: use rcu_dereference in in_vfork()
Arnd Bergmann <[email protected]>
stop_machine: mark helpers __always_inline
Anna-Maria Behnsen <[email protected]>
hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event()
Ard Biesheuvel <[email protected]>
arm64: mm: use a 48-bit ID map when possible on 52-bit VA builds
Daiyue Zhang <[email protected]>
configfs: fix a use-after-free in __configfs_open_file
Jia-Ju Bai <[email protected]>
block: rsxx: fix error return code of rsxx_pci_probe()
Ondrej Mosnacek <[email protected]>
NFSv4.2: fix return value of _nfs4_get_security_label()
Trond Myklebust <[email protected]>
NFS: Don't gratuitously clear the inode cache when lookup failed
Trond Myklebust <[email protected]>
NFS: Don't revalidate the directory permissions on a lookup failure
Benjamin Coddington <[email protected]>
SUNRPC: Set memalloc_nofs_save() for sync tasks
Anshuman Khandual <[email protected]>
arm64/mm: Fix pfn_valid() for ZONE_DEVICE based memory
Sergey Shtylyov <[email protected]>
sh_eth: fix TRSCER mask for R7S72100
Ian Abbott <[email protected]>
staging: comedi: pcl818: Fix endian problem for AI command data
Ian Abbott <[email protected]>
staging: comedi: pcl711: Fix endian problem for AI command data
Ian Abbott <[email protected]>
staging: comedi: me4000: Fix endian problem for AI command data
Ian Abbott <[email protected]>
staging: comedi: dmm32at: Fix endian problem for AI command data
Ian Abbott <[email protected]>
staging: comedi: das800: Fix endian problem for AI command data
Ian Abbott <[email protected]>
staging: comedi: das6402: Fix endian problem for AI command data
Ian Abbott <[email protected]>
staging: comedi: adv_pci1710: Fix endian problem for AI command data
Ian Abbott <[email protected]>
staging: comedi: addi_apci_1500: Fix endian problem for command sample
Ian Abbott <[email protected]>
staging: comedi: addi_apci_1032: Fix endian problem for COS sample
Lee Gibson <[email protected]>
staging: rtl8192e: Fix possible buffer overflow in _rtl92e_wx_set_scan
Lee Gibson <[email protected]>
staging: rtl8712: Fix possible buffer overflow in r8712_sitesurvey_cmd
Dan Carpenter <[email protected]>
staging: ks7010: prevent buffer overflow in ks_wlan_set_scan()
Dan Carpenter <[email protected]>
staging: rtl8188eu: fix potential memory corruption in rtw_check_beacon_data()
Dan Carpenter <[email protected]>
staging: rtl8712: unterminated string leads to read overflow
Dan Carpenter <[email protected]>
staging: rtl8188eu: prevent ->ssid overflow in rtw_wx_set_scan()
Dan Carpenter <[email protected]>
staging: rtl8192u: fix ->ssid overflow in r8192_wx_set_scan()
Dmitry Baryshkov <[email protected]>
misc: fastrpc: restrict user apps from sending kernel RPC messages
Shile Zhang <[email protected]>
misc/pvpanic: Export module FDT device table
Shuah Khan <[email protected]>
usbip: fix vudc usbip_sockfd_store races leading to gpf
Shuah Khan <[email protected]>
usbip: fix vhci_hcd attach_store() races leading to gpf
Shuah Khan <[email protected]>
usbip: fix stub_dev usbip_sockfd_store() races leading to gpf
Shuah Khan <[email protected]>
usbip: fix vudc to check for stream socket
Shuah Khan <[email protected]>
usbip: fix vhci_hcd to check for stream socket
Shuah Khan <[email protected]>
usbip: fix stub_dev to check for stream socket
Sebastian Reichel <[email protected]>
USB: serial: cp210x: add some more GE USB IDs
Karan Singhal <[email protected]>
USB: serial: cp210x: add ID for Acuity Brands nLight Air Adapter
Niv Sardi <[email protected]>
USB: serial: ch341: add new Product ID
Pavel Skripkin <[email protected]>
USB: serial: io_edgeport: fix memory leak in edge_startup
Mathias Nyman <[email protected]>
xhci: Fix repeated xhci wake after suspend due to uncleared internal wake state
Forest Crossman <[email protected]>
usb: xhci: Fix ASMedia ASM1042A and ASM3242 DMA addressing
Mathias Nyman <[email protected]>
xhci: Improve detection of device initiated wake signal.
Stanislaw Gruszka <[email protected]>
usb: xhci: do not perform Soft Retry for some xHCI hosts
Yoshihiro Shimoda <[email protected]>
usb: renesas_usbhs: Clear PIPECFG for re-enabling pipe with other EPNUM
Pete Zaitcev <[email protected]>
USB: usblp: fix a hang in poll() if disconnected
Matthias Kaehlcke <[email protected]>
usb: dwc3: qcom: Honor wakeup enabled/disabled state
Serge Semin <[email protected]>
usb: dwc3: qcom: Add missing DWC3 OF node refcount decrement
Ruslan Bilovol <[email protected]>
usb: gadget: f_uac1: stop playback on function disable
Ruslan Bilovol <[email protected]>
usb: gadget: f_uac2: always increase endpoint max_packet_size by one audio slot
Dan Carpenter <[email protected]>
USB: gadget: u_ether: Fix a configfs return code
Yorick de Wid <[email protected]>
Goodix Fingerprint device is not a modem
Frank Li <[email protected]>
mmc: cqhci: Fix random crash when remove mmc module/card
Adrian Hunter <[email protected]>
mmc: core: Fix partition switch time for eMMC
Heikki Krogerus <[email protected]>
software node: Fix node registration
Stefan Haberland <[email protected]>
s390/dasd: fix hanging IO request during DASD driver unbind
Stefan Haberland <[email protected]>
s390/dasd: fix hanging DASD driver unbind
Andrey Konovalov <[email protected]>
arm64: kasan: fix page_alloc tagging with DEBUG_VIRTUAL
Eric W. Biederman <[email protected]>
Revert 95ebabde382c ("capabilities: Don't allow writing ambiguous v3 file capabilities")
Takashi Iwai <[email protected]>
ALSA: usb-audio: Apply the control quirk to Plantronics headsets
Takashi Iwai <[email protected]>
ALSA: usb-audio: Fix "cannot get freq eq" errors on Dell AE515 sound bar
Takashi Iwai <[email protected]>
ALSA: hda: Avoid spurious unsol event handling during S3/S4
Takashi Iwai <[email protected]>
ALSA: hda: Flush pending unsolicited events before suspend
Takashi Iwai <[email protected]>
ALSA: hda: Drop the BATCH workaround for AMD controllers
Simeon Simeonoff <[email protected]>
ALSA: hda/ca0132: Add Sound BlasterX AE-5 Plus support
Takashi Iwai <[email protected]>
ALSA: hda/hdmi: Cancel pending works before suspend
John Ernberg <[email protected]>
ALSA: usb: Add Plantronics C320-M USB ctrl msg delay quirk
Aleksandr Miloserdov <[email protected]>
scsi: target: core: Prevent underflow for service actions
Aleksandr Miloserdov <[email protected]>
scsi: target: core: Add cmd length set before cmd complete
Mike Christie <[email protected]>
scsi: libiscsi: Fix iscsi_prep_scsi_cmd_pdu() error handling
Lin Feng <[email protected]>
sysctl.c: fix underflow value setting risk in vm_table
Heiko Carstens <[email protected]>
s390/smp: __smp_rescan_cpus() - move cpumask away from stack
Keita Suzuki <[email protected]>
i40e: Fix memory leak in i40e_probe
Geert Uytterhoeven <[email protected]>
PCI: Fix pci_register_io_range() memory leak
Sasha Levin <[email protected]>
kbuild: clamp SUBLEVEL to 255
Krzysztof Wilczyński <[email protected]>
PCI: mediatek: Add missing of_node_put() to fix reference leak
Martin Kaiser <[email protected]>
PCI: xgene-msi: Fix race in installing chained irq handler
Ronald Tschalär <[email protected]>
Input: applespi - don't wait for responses to commands indefinitely.
Khalid Aziz <[email protected]>
sparc64: Use arch_validate_flags() to validate ADI flag
Andreas Larsson <[email protected]>
sparc32: Limit memblock allocation to low memory
Suravee Suthikulpanit <[email protected]>
iommu/amd: Fix performance counter initialization
Michael Ellerman <[email protected]>
powerpc/64: Fix stack trace not displaying final frame
Filipe Laíns <[email protected]>
HID: logitech-dj: add support for the new lightspeed connection iteration
Athira Rajeev <[email protected]>
powerpc/perf: Record counter overflow always if SAMPLE_IP is unset
Nicholas Piggin <[email protected]>
powerpc: improve handling of unrecoverable system reset
Alain Volmat <[email protected]>
spi: stm32: make spurious and overrun interrupts visible
Oliver O'Halloran <[email protected]>
powerpc/pci: Add ppc_md.discover_phbs()
Lubomir Rintel <[email protected]>
Platform: OLPC: Fix probe error handling
Chaotian Jing <[email protected]>
mmc: mediatek: fix race condition between msdc_request_timeout and irq
Christophe JAILLET <[email protected]>
mmc: mxs-mmc: Fix a resource leak in an error handling path in 'mxs_mmc_probe()'
Steven J. Magnani <[email protected]>
udf: fix silent AED tagLocation corruption
Wolfram Sang <[email protected]>
i2c: rcar: optimize cacheline to minimize HW race condition
Wolfram Sang <[email protected]>
i2c: rcar: faster irq code to minimize HW race condition
Guangbin Huang <[email protected]>
net: phy: fix save wrong speed and duplex problem if autoneg is on
Vladimir Oltean <[email protected]>
net: enetc: initialize RFS/RSS memories for unused ports too
Jian Shen <[email protected]>
net: hns3: fix error mask definition of flow director
Hans Verkuil <[email protected]>
media: rc: compile rc-cec.c into rc-core
Biju Das <[email protected]>
media: v4l: vsp1: Fix bru null pointer access
Biju Das <[email protected]>
media: v4l: vsp1: Fix uif null pointer access
Maxim Mikityanskiy <[email protected]>
media: usbtv: Fix deadlock on suspend
Sergey Shtylyov <[email protected]>
sh_eth: fix TRSCER mask for R7S9210
Colin Ian King <[email protected]>
qxl: Fix uninitialised struct field head.surface_id
Wang Qing <[email protected]>
s390/crypto: return -EFAULT if copy_to_user() fails
Eric Farman <[email protected]>
s390/cio: return -EFAULT if copy_to_user() fails
Artem Lapkin <[email protected]>
drm: meson_drv add shutdown function
Neil Roberts <[email protected]>
drm/shmem-helper: Don't remove the offset in vm_area_struct pgoff
Neil Roberts <[email protected]>
drm/shmem-helper: Check for purged buffers in fault handler
Daniel Vetter <[email protected]>
drm/compat: Clear bounce structures
Edwin Peer <[email protected]>
bnxt_en: reliably allocate IRQ table on reset to avoid crash
Wang Qing <[email protected]>
s390/cio: return -EFAULT if copy_to_user() fails again
Jian Shen <[email protected]>
net: hns3: fix bug when calculating the TCAM table info
Jian Shen <[email protected]>
net: hns3: fix query vlan mask value error for flow director
Ian Rogers <[email protected]>
perf traceevent: Ensure read cmdlines are null terminated.
Danielle Ratson <[email protected]>
selftests: forwarding: Fix race condition in mirror installation
Joakim Zhang <[email protected]>
net: stmmac: fix watchdog timeout during suspend/resume stress test
Joakim Zhang <[email protected]>
net: stmmac: stop each tx channel independently
Antony Antony <[email protected]>
ixgbe: fail to create xfrm offload of IPsec tunnel mode SA
Jia-Ju Bai <[email protected]>
net: qrtr: fix error return code of qrtr_sendmsg()
Paul Cercueil <[email protected]>
net: davicom: Fix regulator not turned off on driver removal
Paul Cercueil <[email protected]>
net: davicom: Fix regulator not turned off on failed probe
Xie He <[email protected]>
net: lapbether: Remove netif_start_queue / netif_stop_queue
Paul Moore <[email protected]>
cipso,calipso: resolve a number of problems with the DOI refcounts
Hillf Danton <[email protected]>
netdevsim: init u64 stats for 32bit hardware
Daniele Palmas <[email protected]>
net: usb: qmi_wwan: allow qmimux add/del with master up
Maximilian Heyne <[email protected]>
net: sched: avoid duplicates in classes dump
Ido Schimmel <[email protected]>
nexthop: Do not flush blackhole nexthops when loopback goes down
Ong Boon Leong <[email protected]>
net: stmmac: fix incorrect DMA channel intr enable setting of EQoS v4.10
Kevin(Yudong) Yang <[email protected]>
net/mlx4_en: update moderation when config reset
Vladimir Oltean <[email protected]>
net: enetc: don't overwrite the RSS indirection table when initializing
Linus Torvalds <[email protected]>
Revert "mm, slub: consider rest of partial list if acquire_slab() fails"
Paulo Alcantara <[email protected]>
cifs: return proper error code in statfs(2)
Christian Brauner <[email protected]>
mount: fix mounting of detached mounts onto targets that reside on shared mounts
Christophe Leroy <[email protected]>
powerpc/603: Fix protection of user pages mapped with PROT_NONE
Lorenzo Bianconi <[email protected]>
mt76: dma: do not report truncated frames to mac80211
Jiri Wiesner <[email protected]>
ibmvnic: always store valid MAC address
Maciej Fijalkowski <[email protected]>
samples, bpf: Add missing munmap in xdpsock
Yauheni Kaliuta <[email protected]>
selftests/bpf: Mask bpf_csum_diff() return value to 16 bits in test_verifier
Hangbin Liu <[email protected]>
selftests/bpf: No need to drop the packet when there is no geneve opt
Vasily Averin <[email protected]>
netfilter: x_tables: gpf inside xt_find_revision()
Florian Westphal <[email protected]>
netfilter: nf_nat: undo erroneous tcp edemux lookup
Eric Dumazet <[email protected]>
tcp: add sanity tests to TCP_QUEUE_SEQ
Torin Cooper-Bennun <[email protected]>
can: tcan4x5x: tcan4x5x_init(): fix initialization - clear MRAM before entering Normal Mode
Joakim Zhang <[email protected]>
can: flexcan: invoke flexcan_chip_freeze() to enter freeze mode
Joakim Zhang <[email protected]>
can: flexcan: enable RX FIFO after FRZ/HALT valid
Joakim Zhang <[email protected]>
can: flexcan: assert FRZ bit in flexcan_chip_freeze()
Oleksij Rempel <[email protected]>
can: skb: can_skb_set_owner(): fix ref counting if socket was closed before setting skb ownership
Sergey Shtylyov <[email protected]>
sh_eth: fix TRSCER mask for SH771x
Balazs Nemeth <[email protected]>
net: avoid infinite loop in mpls_gso_segment when mpls_hlen == 0
Balazs Nemeth <[email protected]>
net: check if protocol extracted by virtio_net_hdr_set_proto is correct
Daniel Borkmann <[email protected]>
net: Fix gro aggregation for udp encaps with zero csum
Felix Fietkau <[email protected]>
ath9k: fix transmitting to stations in dynamic SMPS mode
Jakub Kicinski <[email protected]>
ethernet: alx: fix order of calls on resume
Greg Kurz <[email protected]>
powerpc/pseries: Don't enforce MSI affinity with kdump
Dmitry V. Levin <[email protected]>
uapi: nfnetlink_cthelper.h: fix userspace compilation error
-------------
Diffstat:
Documentation/virt/kvm/api.txt | 3 + Makefile | 16 ++- arch/arm/include/asm/kvm_asm.h | 2 +- arch/arm/kvm/hyp/tlb.c | 3 +- arch/arm64/include/asm/kvm_asm.h | 2 +- arch/arm64/include/asm/memory.h | 5 + arch/arm64/include/asm/mmu_context.h | 5 +- arch/arm64/kernel/head.S | 2 +- arch/arm64/kvm/hyp/tlb.c | 3 +- arch/arm64/kvm/reset.c | 11 +- arch/arm64/mm/init.c | 12 ++ arch/arm64/mm/mmu.c | 2 +- arch/powerpc/include/asm/code-patching.h | 2 +- arch/powerpc/include/asm/machdep.h | 3 + arch/powerpc/include/asm/ptrace.h | 3 + arch/powerpc/kernel/asm-offsets.c | 2 +- arch/powerpc/kernel/head_32.S | 9 +- arch/powerpc/kernel/pci-common.c | 10 ++ arch/powerpc/kernel/process.c | 2 +- arch/powerpc/kernel/traps.c | 5 +- arch/powerpc/perf/core-book3s.c | 19 ++- arch/powerpc/platforms/pseries/msi.c | 25 +++- arch/s390/kernel/smp.c | 2 +- arch/sparc/include/asm/mman.h | 54 +++++---- arch/sparc/mm/init_32.c | 3 + arch/x86/kernel/unwind_orc.c | 12 +- drivers/base/swnode.c | 3 + drivers/block/rsxx/core.c | 1 + drivers/block/zram/zram_drv.c | 11 +- drivers/gpu/drm/drm_gem_shmem_helper.c | 25 ++-- drivers/gpu/drm/drm_ioc32.c | 11 ++ drivers/gpu/drm/meson/meson_drv.c | 11 ++ drivers/gpu/drm/qxl/qxl_display.c | 1 + drivers/hid/hid-logitech-dj.c | 7 +- drivers/i2c/busses/i2c-rcar.c | 13 +- drivers/input/keyboard/applespi.c | 21 +++- drivers/iommu/amd_iommu_init.c | 45 +++++-- drivers/media/platform/vsp1/vsp1_drm.c | 6 +- drivers/media/rc/Makefile | 1 + drivers/media/rc/keymaps/Makefile | 1 - drivers/media/rc/keymaps/rc-cec.c | 28 ++--- drivers/media/rc/rc-main.c | 6 + drivers/media/usb/usbtv/usbtv-audio.c | 2 +- drivers/misc/fastrpc.c | 5 + drivers/misc/pvpanic.c | 1 + drivers/mmc/core/bus.c | 11 +- drivers/mmc/core/mmc.c | 15 ++- drivers/mmc/host/mtk-sd.c | 18 +-- drivers/mmc/host/mxs-mmc.c | 2 +- drivers/net/can/flexcan.c | 24 ++-- drivers/net/can/m_can/tcan4x5x.c | 6 +- drivers/net/ethernet/atheros/alx/main.c | 7 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 14 ++- drivers/net/ethernet/davicom/dm9000.c | 21 +++- drivers/net/ethernet/freescale/enetc/enetc.c | 19 ++- drivers/net/ethernet/freescale/enetc/enetc.h | 5 + drivers/net/ethernet/freescale/enetc/enetc_pf.c | 40 ++++++- drivers/net/ethernet/freescale/enetc/enetc_vf.c | 7 ++ .../net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h | 6 +- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 7 +- drivers/net/ethernet/ibm/ibmvnic.c | 5 +- drivers/net/ethernet/intel/i40e/i40e_main.c | 2 + drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c | 5 + drivers/net/ethernet/intel/ixgbevf/ipsec.c | 5 + drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 2 +- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 2 + drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 1 + drivers/net/ethernet/renesas/sh_eth.c | 7 ++ drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c | 19 ++- drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c | 4 - drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 + drivers/net/netdevsim/netdev.c | 1 + drivers/net/phy/phy.c | 7 +- drivers/net/usb/qmi_wwan.c | 14 --- drivers/net/wan/lapbether.c | 3 - drivers/net/wireless/ath/ath9k/ath9k.h | 3 +- drivers/net/wireless/ath/ath9k/xmit.c | 6 + drivers/net/wireless/mediatek/mt76/dma.c | 11 +- drivers/nvme/host/core.c | 8 +- drivers/pci/controller/pci-xgene-msi.c | 10 +- drivers/pci/controller/pcie-mediatek.c | 7 +- drivers/pci/pci.c | 4 + drivers/platform/olpc/olpc-ec.c | 15 +-- drivers/s390/block/dasd.c | 6 +- drivers/s390/cio/vfio_ccw_ops.c | 6 +- drivers/s390/crypto/vfio_ap_ops.c | 2 +- drivers/scsi/libiscsi.c | 11 +- drivers/spi/spi-stm32.c | 15 +-- drivers/staging/comedi/drivers/addi_apci_1032.c | 4 +- drivers/staging/comedi/drivers/addi_apci_1500.c | 18 +-- drivers/staging/comedi/drivers/adv_pci1710.c | 10 +- drivers/staging/comedi/drivers/das6402.c | 2 +- drivers/staging/comedi/drivers/das800.c | 2 +- drivers/staging/comedi/drivers/dmm32at.c | 2 +- drivers/staging/comedi/drivers/me4000.c | 2 +- drivers/staging/comedi/drivers/pcl711.c | 2 +- drivers/staging/comedi/drivers/pcl818.c | 2 +- drivers/staging/ks7010/ks_wlan_net.c | 6 +- drivers/staging/rtl8188eu/core/rtw_ap.c | 5 + drivers/staging/rtl8188eu/os_dep/ioctl_linux.c | 6 +- drivers/staging/rtl8192e/rtl8192e/rtl_wx.c | 7 +- drivers/staging/rtl8192u/r8192U_wx.c | 6 +- drivers/staging/rtl8712/rtl871x_cmd.c | 6 +- drivers/staging/rtl8712/rtl871x_ioctl_linux.c | 2 +- drivers/target/target_core_pr.c | 15 ++- drivers/target/target_core_transport.c | 15 ++- drivers/usb/class/cdc-acm.c | 5 + drivers/usb/class/usblp.c | 16 ++- drivers/usb/dwc3/dwc3-qcom.c | 16 ++- drivers/usb/gadget/function/f_uac1.c | 1 + drivers/usb/gadget/function/f_uac2.c | 2 +- drivers/usb/gadget/function/u_ether_configfs.h | 5 +- drivers/usb/host/xhci-pci.c | 13 +- drivers/usb/host/xhci-ring.c | 3 +- drivers/usb/host/xhci.c | 78 ++++++------ drivers/usb/host/xhci.h | 1 + drivers/usb/renesas_usbhs/pipe.c | 2 + drivers/usb/serial/ch341.c | 1 + drivers/usb/serial/cp210x.c | 3 + drivers/usb/serial/io_edgeport.c | 26 ++-- drivers/usb/usbip/stub_dev.c | 42 +++++-- drivers/usb/usbip/vhci_sysfs.c | 39 +++++- drivers/usb/usbip/vudc_sysfs.c | 49 ++++++-- drivers/xen/events/events_2l.c | 22 ++-- drivers/xen/events/events_base.c | 132 +++++++++++++++------ drivers/xen/events/events_fifo.c | 7 -- drivers/xen/events/events_internal.h | 22 ++-- fs/binfmt_misc.c | 29 +++-- fs/cifs/cifsfs.c | 2 +- fs/configfs/file.c | 6 +- fs/nfs/dir.c | 40 ++++--- fs/nfs/nfs4proc.c | 2 +- fs/pnode.h | 2 +- fs/udf/inode.c | 9 +- include/linux/can/skb.h | 8 +- include/linux/sched/mm.h | 3 +- include/linux/stop_machine.h | 11 +- include/linux/virtio_net.h | 7 +- include/media/rc-map.h | 7 ++ include/target/target_core_backend.h | 1 + include/uapi/linux/netfilter/nfnetlink_cthelper.h | 2 +- kernel/sched/membarrier.c | 4 +- kernel/sys.c | 2 +- kernel/sysctl.c | 8 +- kernel/time/hrtimer.c | 60 ++++++---- lib/logic_pio.c | 3 + mm/slub.c | 2 +- net/ipv4/cipso_ipv4.c | 11 +- net/ipv4/nexthop.c | 10 +- net/ipv4/tcp.c | 23 ++-- net/ipv4/udp_offload.c | 2 +- net/ipv6/calipso.c | 14 +-- net/mpls/mpls_gso.c | 3 + net/netfilter/nf_nat_proto.c | 25 +++- net/netfilter/x_tables.c | 6 +- net/netlabel/netlabel_cipso_v4.c | 3 + net/qrtr/qrtr.c | 4 +- net/sched/sch_api.c | 8 +- net/sunrpc/sched.c | 5 +- samples/bpf/xdpsock_user.c | 2 + security/commoncap.c | 12 +- sound/pci/hda/hda_bind.c | 4 + sound/pci/hda/hda_controller.c | 7 -- sound/pci/hda/hda_intel.c | 2 + sound/pci/hda/patch_ca0132.c | 1 + sound/pci/hda/patch_hdmi.c | 13 ++ sound/usb/quirks.c | 9 ++ tools/perf/util/trace-event-read.c | 1 + .../testing/selftests/bpf/progs/test_tunnel_kern.c | 6 +- .../testing/selftests/bpf/verifier/array_access.c | 3 +- .../net/forwarding/mirror_gre_bridge_1d_vlan.sh | 9 ++ virt/kvm/arm/arm.c | 8 +- virt/kvm/arm/mmu.c | 3 +- 173 files changed, 1213 insertions(+), 587 deletions(-)
[email protected]
maintainerPATCH 1/168 uapi: nfnetlink_cthelper.h: fix userspace compilation error
From: Greg Kroah-Hartman <[email protected]>
From: Dmitry V. Levin <[email protected]>
commit c33cb0020ee6dd96cc9976d6085a7d8422f6dbed upstream.
Apparently, <linux/netfilter/nfnetlink_cthelper.h> and <linux/netfilter/nfnetlink_acct.h> could not be included into the same compilation unit because of a cut-and-paste typo in the former header.
--- include/uapi/linux/netfilter/nfnetlink_cthelper.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/uapi/linux/netfilter/nfnetlink_cthelper.h +1 -1
@@ -5,7 +5,7 @@#define NFCT_HELPER_STATUS_DISABLED 0#define NFCT_HELPER_STATUS_ENABLED 1 -enum nfnl_acct_msg_types { +enum nfnl_cthelper_msg_types {NFNL_MSG_CTHELPER_NEW ,NFNL_MSG_CTHELPER_GET ,NFNL_MSG_CTHELPER_DEL ,
[email protected]
maintainerPATCH 2/168 powerpc/pseries: Dont enforce MSI affinity with kdump
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Greg Kurz <[email protected]>
commit f9619d5e5174867536b7e558683bc4408eab833f upstream.
Depending on the number of online CPUs in the original kernel, it is likely for CPU #0 to be offline in a kdump kernel. The associated IRQs in the affinity mappings provided by irq_create_affinity_masks() are thus not started by irq_startup(), as per-design with managed IRQs.
This can be a problem with multi-queue block devices driven by blk-mq :
such a non-started IRQ is very likely paired with the single queue enforced by blk-mq during kdump (see blk_mq_alloc_tag_set()). This causes the device to remain silent and likely hangs the guest at some point.
This is a regression caused by commit 9ea69a55b3b9 ("powerpc/pseries:
Pass MSI affinity to irq_create_mapping()"). Note that this only happens with the XIVE interrupt controller because XICS has a workaround to bypass affinity, which is activated during kdump with the "noirqdistrib" kernel parameter.
The issue comes from a combination of factors:
- discrepancy between the number of queues detected by the multi-queue block driver, that was used to create the MSI vectors, and the single queue mode enforced later on by blk-mq because of kdump (i.e. keeping all queues fixes the issue)
- CPU#0 offline (i.e. kdump always succeed with CPU#0)
Given that I couldn't reproduce on x86, which seems to always have CPU#0 online even during kdump, I'm not sure where this should be fixed. Hence going for another approach : fine-grained affinity is for performance and we don't really care about that during kdump. Simply revert to the previous working behavior of ignoring affinity masks in this case only.
--- arch/powerpc/platforms/pseries/msi.c | 25 +++++++++++++++++++++++-- 1 file changed, 23 insertions(+), 2 deletions(-)
arch/powerpc/platforms/pseries/msi.c +23 -2
@@ -4,6 +4,7 @@* Copyright 2006-2007 Michael Ellerman, IBM Corp. */ +#include <linux/crash_dump.h> #include <linux/device.h> #include <linux/irq.h> #include <linux/msi.h>
@@ -458,8 +459,28 @@ again:return hwirq; } - virq =irq_create_mapping_affinity (NULL , hwirq, - entry->affinity); +/* +* Depending on the number of online CPUs in the original +* kernel, it is likely for CPU #0 to be offline in a kdump +* kernel. The associated IRQs in the affinity mappings +* provided by irq_create_affinity_masks() are thus not +* started by irq_startup(), as per-design for managed IRQs. +* This can be a problem with multi-queue block devices driven +* by blk-mq : such a non-started IRQ is very likely paired +* with the single queue enforced by blk-mq during kdump (see +* blk_mq_alloc_tag_set()). This causes the device to remain +* silent and likely hangs the guest at some point. +* +* We don't really care for fine-grained affinity when doing +* kdump actually : simply ignore the pre-computed affinity +* masks in this case and let the default mask with all CPUs +* be used when creating the IRQ mappings. +*/ +if (is_kdump_kernel ()) + virq =irq_create_mapping (NULL , hwirq); +else + virq =irq_create_mapping_affinity (NULL , hwirq, + entry->affinity);if (!virq) {pr_debug ("rtas_msi: Failed mapping hwirq %d\n" , hwirq);
[email protected]
maintainerPATCH 3/168 ethernet: alx: fix order of calls on resume
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Jakub Kicinski <[email protected]>
commit a4dcfbc4ee2218abd567d81d795082d8d4afcdf6 upstream.
netif_device_attach() will unpause the queues so we can't call it before __alx_open(). This went undetected until commit b0999223f224 ("alx: add ability to allocate and free alx_napi structures") but now if stack tries to xmit immediately on resume before __alx_open() we'll crash on the NAPI being null:
BUG: kernel NULL pointer dereference, address: 0000000000000198 CPU: 0 PID: 12 Comm: ksoftirqd/0 Tainted: G OE 5.10.0-3-amd64 #1 Debian 5.10.13-1 Hardware name: Gigabyte Technology Co., Ltd. To be filled by O.E.M./H77-D3H, BIOS F15 11/14/2013 RIP: 0010:alx_start_xmit+0x34/0x650 [alx] Code: 41 56 41 55 41 54 55 53 48 83 ec 20 0f b7 57 7c 8b 8e b0
0b 00 00 39 ca 72 06 89 d0 31 d2 f7 f1 89 d2 48 8b 84 df
RSP: 0018:ffffb09240083d28 EFLAGS: 00010297
RAX: 0000000000000000 RBX: ffffa04d80ae7800 RCX: 0000000000000004
RDX: 0000000000000000 RSI: ffffa04d80afa000 RDI: ffffa04e92e92a00
RBP: 0000000000000042 R08: 0000000000000100 R09: ffffa04ea3146700
R10: 0000000000000014 R11: 0000000000000000 R12: ffffa04e92e92100
R13: 0000000000000001 R14: ffffa04e92e92a00 R15: ffffa04e92e92a00
FS: 0000000000000000(0000) GS:ffffa0508f600000(0000) knlGS:0000000000000000 i915 0000:00:02.0: vblank wait timed out on crtc 0
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000198 CR3: 000000004460a001 CR4: 00000000001706f0
Call Trace: dev_hard_start_xmit+0xc7/0x1e0 sch_direct_xmit+0x10f/0x310
--- drivers/net/ethernet/atheros/alx/main.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/net/ethernet/atheros/alx/main.c +5 -2
@@ -1897,13 +1897,16 @@ static int alx_resume(struct device *devif (!netif_running (alx->dev))return 0 ; -netif_device_attach (alx->dev);rtnl_lock (); err =__alx_open (alx, true);rtnl_unlock (); +if (err) +return err; + +netif_device_attach (alx->dev); -return err; +return 0 ; } static SIMPLE_DEV_PM_OPS(alx_pm_ops, alx_suspend, alx_resume);
[email protected]
maintainerPATCH 4/168 ath9k: fix transmitting to stations in dynamic SMPS mode
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Felix Fietkau <[email protected]>
commit 3b9ea7206d7e1fdd7419cbd10badd3b2c80d04b4 upstream.
When transmitting to a receiver in dynamic SMPS mode, all transmissions that use multiple spatial streams need to be sent using CTS-to-self or RTS/CTS to give the receiver's extra chains some time to wake up.
This fixes the tx rate getting stuck at <= MCS7 for some clients, especially Intel ones, which make aggressive use of SMPS.
--- drivers/net/wireless/ath/ath9k/ath9k.h | 3 ++- drivers/net/wireless/ath/ath9k/xmit.c | 6 ++++++ 2 files changed, 8 insertions(+), 1 deletion(-)
drivers/net/wireless/ath/ath9k/ath9k.h +2 -1
@@ -177,7 +177,8 @@ struct ath_frame_info {s8 txq;u8 keyix;u8 rtscts_rate; - u8 retries : 7; + u8 retries : 6; + u8 dyn_smps : 1; u8 baw_tracked : 1;u8 tx_power; enum ath9k_key_type keytype:2;
drivers/net/wireless/ath/ath9k/xmit.c +6 -0
@@ -1271,6 +1271,11 @@ static void ath_buf_set_rate(struct ath_ is_40, is_sgi, is_sp);if (rix <8 && (tx_info->flags &IEEE80211_TX_CTL_STBC )) info->rates[i].RateFlags |=ATH9K_RATESERIES_STBC ; +if (rix >=8 && fi->dyn_smps) { + info->rates[i].RateFlags |= +ATH9K_RATESERIES_RTS_CTS ; + info->flags |=ATH9K_TXDESC_CTSENA ; + } info->txpower[i] =ath_get_rate_txpower (sc, bf, rix, is_40, false);
@@ -2111,6 +2116,7 @@ static void setup_frame_info(struct ieee fi->keyix = an->ps_key; else fi->keyix = ATH9K_TXKEYIX_INVALID; + fi->dyn_smps = sta && sta->smps_mode ==IEEE80211_SMPS_DYNAMIC ; fi->keytype = keytype; fi->framelen = framelen; fi->tx_power = txpower;
[email protected]
maintainerPATCH 5/168 net: Fix gro aggregation for udp encaps with zero csum
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Daniel Borkmann <[email protected]>
commit 89e5c58fc1e2857ccdaae506fb8bc5fed57ee063 upstream.
We noticed a GRO issue for UDP-based encaps such as vxlan/geneve when the csum for the UDP header itself is 0. In that case, GRO aggregation does not take place on the phys dev, but instead is deferred to the vxlan/geneve driver (see trace below).
The reason is essentially that GRO aggregation bails out in udp_gro_receive() for such case when drivers marked the skb with CHECKSUM_UNNECESSARY (ice, i40e, others) where for non-zero csums 2abb7cdc0dc8 ("udp: Add support for doing checksum unnecessary conversion") promotes those skbs to CHECKSUM_COMPLETE and napi context has csum_valid set. This is however not the case for zero UDP csum (here: csum_cnt is still 0 and csum_valid continues to be false).
At the same time 57c67ff4bd92 ("udp: additional GRO support") added matches on !uh->check ^ !uh2->check as part to determine candidates for aggregation, so it certainly is expected to handle zero csums in udp_gro_receive(). The purpose of the check added via 662880f44203 ("net: Allow GRO to use and set levels of checksum unnecessary") seems to catch bad csum and stop aggregation right away.
One way to fix aggregation in the zero case is to only perform the !csum_valid check in udp_gro_receive() if uh->check is infact non-zero.
Before:
[...]
swapper 0 [008] 731.946506: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100400 len=1500 (1) swapper 0 [008] 731.946507: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100200 len=1500 swapper 0 [008] 731.946507: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101100 len=1500 swapper 0 [008] 731.946508: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101700 len=1500 swapper 0 [008] 731.946508: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101b00 len=1500 swapper 0 [008] 731.946508: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100600 len=1500 swapper 0 [008] 731.946508: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100f00 len=1500 swapper 0 [008] 731.946509: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100a00 len=1500 swapper 0 [008] 731.946516: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100500 len=1500 swapper 0 [008] 731.946516: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100700 len=1500 swapper 0 [008] 731.946516: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101d00 len=1500 (2) swapper 0 [008] 731.946517: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101000 len=1500 swapper 0 [008] 731.946517: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101c00 len=1500 swapper 0 [008] 731.946517: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101400 len=1500 swapper 0 [008] 731.946518: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100e00 len=1500 swapper 0 [008] 731.946518: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497101600 len=1500 swapper 0 [008] 731.946521: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff966497100800 len=774 swapper 0 [008] 731.946530: net:netif_receive_skb: dev=test_vxlan skbaddr=0xffff966497100400 len=14032 (1) swapper 0 [008] 731.946530: net:netif_receive_skb: dev=test_vxlan skbaddr=0xffff966497101d00 len=9112 (2)
[...]
# netperf -H 10.55.10.4 -t TCP_STREAM -l 20 MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.55.10.4 () port 0 AF_INET : demo Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 20.01 13129.24
After:
[...]
swapper 0 [026] 521.862641: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff93ab0d479000 len=11286 (1) swapper 0 [026] 521.862643: net:netif_receive_skb: dev=test_vxlan skbaddr=0xffff93ab0d479000 len=11236 (1) swapper 0 [026] 521.862650: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff93ab0d478500 len=2898 (2) swapper 0 [026] 521.862650: net:netif_receive_skb: dev=enp10s0f0 skbaddr=0xffff93ab0d479f00 len=8490 (3) swapper 0 [026] 521.862653: net:netif_receive_skb: dev=test_vxlan skbaddr=0xffff93ab0d478500 len=2848 (2) swapper 0 [026] 521.862653: net:netif_receive_skb: dev=test_vxlan skbaddr=0xffff93ab0d479f00 len=8440 (3)
[...]
# netperf -H 10.55.10.4 -t TCP_STREAM -l 20 MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.55.10.4 () port 0 AF_INET : demo Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 20.01 24576.53
--- net/ipv4/udp_offload.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/udp_offload.c +1 -1
@@ -426,7 +426,7 @@ struct sk_buff *udp_gro_receive(struct l }if (NAPI_GRO_CB (skb)->encap_mark || - (skb->ip_summed !=CHECKSUM_PARTIAL && + (uh->check && skb->ip_summed !=CHECKSUM_PARTIAL &&NAPI_GRO_CB (skb)->csum_cnt ==0 && !NAPI_GRO_CB (skb)->csum_valid) || !udp_sk (sk)->gro_receive)
[email protected]
maintainerPATCH 6/168 net: check if protocol extracted by virtio_net_hdr_set_proto is correct
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Balazs Nemeth <[email protected]>
commit 924a9bc362a5223cd448ca08c3dde21235adc310 upstream.
For gso packets, virtio_net_hdr_set_proto sets the protocol (if it isn't set) based on the type in the virtio net hdr, but the skb could contain anything since it could come from packet_snd through a raw socket. If there is a mismatch between what virtio_net_hdr_set_proto sets and the actual protocol, then the skb could be handled incorrectly later on.
An example where this poses an issue is with the subsequent call to skb_flow_dissect_flow_keys_basic which relies on skb->protocol being set correctly. A specially crafted packet could fool skb_flow_dissect_flow_keys_basic preventing EINVAL to be returned.
Avoid blindly trusting the information provided by the virtio net header by checking that the protocol in the packet actually matches the protocol set by virtio_net_hdr_set_proto. Note that since the protocol is only checked if skb->dev implements header_ops->parse_protocol, packets from devices without the implementation are not checked at this stage.
--- include/linux/virtio_net.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
include/linux/virtio_net.h +6 -1
@@ -79,8 +79,13 @@ static inline int virtio_net_hdr_to_skb( if (gso_type && skb->network_header) { struct flow_keys_basic keys; - if (!skb->protocol) + if (!skb->protocol) { + __be16 protocol = dev_parse_header_protocol(skb); + virtio_net_hdr_set_proto(skb, hdr); + if (protocol && protocol != skb->protocol) + return -EINVAL; + } retry: if (!skb_flow_dissect_flow_keys_basic(NULL, skb, &keys, NULL, 0, 0, 0,
[email protected]
maintainerPATCH 7/168 net: avoid infinite loop in mpls_gso_segment when mpls_hlen == 0
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Balazs Nemeth <[email protected]>
commit d348ede32e99d3a04863e9f9b28d224456118c27 upstream.
A packet with skb_inner_network_header(skb) == skb_network_header(skb) and ETH_P_MPLS_UC will prevent mpls_gso_segment from pulling any headers from the packet. Subsequently, the call to skb_mac_gso_segment will again call mpls_gso_segment with the same packet leading to an infinite loop. In addition, ensure that the header length is a multiple of four, which should hold irrespective of the number of stacked labels.
--- net/mpls/mpls_gso.c | 3 +++ 1 file changed, 3 insertions(+)
net/mpls/mpls_gso.c +3 -0
@@ -14,6 +14,7 @@#include <linux/netdev_features.h> #include <linux/netdevice.h> #include <linux/skbuff.h> +#include <net/mpls.h> static struct sk_buff *mpls_gso_segment (struct sk_buff *skb,netdev_features_t features)
@@ -27,6 +28,8 @@ static struct sk_buff *mpls_gso_segment(skb_reset_network_header (skb); mpls_hlen =skb_inner_network_header (skb) -skb_network_header (skb); +if (unlikely (!mpls_hlen || mpls_hlen %MPLS_HLEN )) + goto out;if (unlikely (!pskb_may_pull (skb, mpls_hlen))) goto out;
[email protected]
maintainerPATCH 8/168 sh_eth: fix TRSCER mask for SH771x
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Sergey Shtylyov <[email protected]>
commit 8c91bc3d44dfef8284af384877fbe61117e8b7d1 upstream.
According to the SH7710, SH7712, SH7713 Group User's Manual: Hardware, Rev. 3.00, the TRSCER register actually has only bit 7 valid (and named differently), with all the other bits reserved. Apparently, this was not the case with some early revisions of the manual as we have the other bits declared (and set) in the original driver. Follow the suit and add the explicit sh_eth_cpu_data::trscer_err_mask initializer for SH771x...
--- drivers/net/ethernet/renesas/sh_eth.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/ethernet/renesas/sh_eth.c +3 -0
@@ -1131,6 +1131,9 @@ static struct sh_eth_cpu_data sh771x_datEESIPR_CEEFIP |EESIPR_CELFIP |EESIPR_RRFIP |EESIPR_RTLFIP |EESIPR_RTSFIP | EESIPR_PREIP | EESIPR_CERFIP, + + .trscer_err_mask = DESC_I_RINT8, + .tsu = 1, .dual_port = 1, };
[email protected]
maintainerPATCH 9/168 can: skb: can_skb_set_owner(): fix ref counting if socket was closed before setting skb ownership
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Oleksij Rempel <[email protected]>
commit e940e0895a82c6fbaa259f2615eb52b57ee91a7e upstream.
There are two ref count variables controlling the free()ing of a socket:
- struct sock::sk_refcnt - which is changed by sock_hold()/sock_put()
- struct sock::sk_wmem_alloc - which accounts the memory allocated by the skbs in the send path.
In case there are still TX skbs on the fly and the socket() is closed, the struct sock::sk_refcnt reaches 0. In the TX-path the CAN stack clones an "echo" skb, calls sock_hold() on the original socket and references it. This produces the following back trace:
| WARNING: CPU: 0 PID: 280 at lib/refcount.c:25 refcount_warn_saturate+0x114/0x134 | refcount_t: addition on 0; use-after-free. | Modules linked in: coda_vpu(E) v4l2_jpeg(E) videobuf2_vmalloc(E) imx_vdoa(E) | CPU: 0 PID: 280 Comm: test_can.sh Tainted: G E 5.11.0-04577-gf8ff6603c617 #203 | Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree) | Backtrace: | [<80bafea4>] (dump_backtrace) from [<80bb0280>] (show_stack+0x20/0x24) r7:00000000 r6:600f0113 r5:00000000 r4:81441220 | [<80bb0260>] (show_stack) from [<80bb593c>] (dump_stack+0xa0/0xc8) | [<80bb589c>] (dump_stack) from [<8012b268>] (__warn+0xd4/0x114) r9:00000019 r8:80f4a8c2 r7:83e4150c r6:00000000 r5:00000009 r4:80528f90 | [<8012b194>] (__warn) from [<80bb09c4>] (warn_slowpath_fmt+0x88/0xc8) r9:83f26400 r8:80f4a8d1 r7:00000009 r6:80528f90 r5:00000019 r4:80f4a8c2 | [<80bb0940>] (warn_slowpath_fmt) from [<80528f90>] (refcount_warn_saturate+0x114/0x134) r8:00000000 r7:00000000 r6:82b44000 r5:834e5600 r4:83f4d540 | [<80528e7c>] (refcount_warn_saturate) from [<8079a4c8>] (__refcount_add.constprop.0+0x4c/0x50) | [<8079a47c>] (__refcount_add.constprop.0) from [<8079a57c>] (can_put_echo_skb+0xb0/0x13c) | [<8079a4cc>] (can_put_echo_skb) from [<8079ba98>] (flexcan_start_xmit+0x1c4/0x230) r9:00000010 r8:83f48610 r7:0fdc0000 r6:0c080000 r5:82b44000 r4:834e5600 | [<8079b8d4>] (flexcan_start_xmit) from [<80969078>] (netdev_start_xmit+0x44/0x70) r9:814c0ba0 r8:80c8790c r7:00000000 r6:834e5600 r5:82b44000 r4:82ab1f00 | [<80969034>] (netdev_start_xmit) from [<809725a4>] (dev_hard_start_xmit+0x19c/0x318) r9:814c0ba0 r8:00000000 r7:82ab1f00 r6:82b44000 r5:00000000 r4:834e5600 | [<80972408>] (dev_hard_start_xmit) from [<809c6584>] (sch_direct_xmit+0xcc/0x264) r10:834e5600 r9:00000000 r8:00000000 r7:82b44000 r6:82ab1f00 r5:834e5600 r4:83f27400 | [<809c64b8>] (sch_direct_xmit) from [<809c6c0c>] (__qdisc_run+0x4f0/0x534)
To fix this problem, only set skb ownership to sockets which have still a ref count > 0.
--- include/linux/can/skb.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
include/linux/can/skb.h +6 -2
@@ -49,8 +49,12 @@ static inline void can_skb_reserve(strucstatic inline void can_skb_set_owner (struct sk_buff *skb,struct sock *sk) { -if (sk) { -sock_hold (sk); +/* If the socket has already been closed by user space, the +* refcount may already be 0 (and the socket will be freed +* after the last TX skb has been freed). So only increase +* socket refcount if the refcount is > 0. +*/ +if (sk &&refcount_inc_not_zero (&sk->sk_refcnt)) { skb->destructor = sock_efree; skb->sk = sk; }
[email protected]
maintainerPATCH 10/168 can: flexcan: assert FRZ bit in flexcan_chip_freeze()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Joakim Zhang <[email protected]>
commit 449052cfebf624b670faa040245d3feed770d22f upstream.
Assert HALT bit to enter freeze mode, there is a premise that FRZ bit is asserted. This patch asserts FRZ bit in flexcan_chip_freeze, although the reset value is 1b'1. This is a prepare patch, later patch will invoke flexcan_chip_freeze() to enter freeze mode, which polling freeze mode acknowledge.
--- drivers/net/can/flexcan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/can/flexcan.c +1 -1
@@ -548,7 +548,7 @@ static int flexcan_chip_freeze(struct flu32 reg; reg = priv->read (®s->mcr); - reg |=FLEXCAN_MCR_HALT ; + reg |=FLEXCAN_MCR_FRZ |FLEXCAN_MCR_HALT ; priv->write (reg, ®s->mcr);while (timeout-- && !(priv->read (®s->mcr) &FLEXCAN_MCR_FRZ_ACK ))
[email protected]
maintainerPATCH 11/168 can: flexcan: enable RX FIFO after FRZ/HALT valid
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Joakim Zhang <[email protected]>
commit ec15e27cc8904605846a354bb1f808ea1432f853 upstream.
RX FIFO enable failed could happen when do system reboot stress test:
[ 0.303958] flexcan 5a8d0000.can: 5a8d0000.can supply xceiver not found, using dummy regulator [ 0.304281] flexcan 5a8d0000.can (unnamed net_device) (uninitialized): Could not enable RX FIFO, unsupported core [ 0.314640] flexcan 5a8d0000.can: registering netdev failed [ 0.320728] flexcan 5a8e0000.can: 5a8e0000.can supply xceiver not found, using dummy regulator [ 0.320991] flexcan 5a8e0000.can (unnamed net_device) (uninitialized): Could not enable RX FIFO, unsupported core [ 0.331360] flexcan 5a8e0000.can: registering netdev failed [ 0.337444] flexcan 5a8f0000.can: 5a8f0000.can supply xceiver not found, using dummy regulator [ 0.337716] flexcan 5a8f0000.can (unnamed net_device) (uninitialized): Could not enable RX FIFO, unsupported core [ 0.348117] flexcan 5a8f0000.can: registering netdev failed
RX FIFO should be enabled after the FRZ/HALT are valid. But the current code enable RX FIFO and FRZ/HALT at the same time.
--- drivers/net/can/flexcan.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/net/can/flexcan.c +7 -3
@@ -1432,10 +1432,14 @@ static int register_flexcandev(struct neif (err) goto out_chip_disable; -/* set freeze, halt and activate FIFO, restrict register access */ +/* set freeze, halt */ + err =flexcan_chip_freeze (priv); +if (err) + goto out_chip_disable; + +/* activate FIFO, restrict register access */ reg = priv->read (®s->mcr); - reg |=FLEXCAN_MCR_FRZ |FLEXCAN_MCR_HALT | -FLEXCAN_MCR_FEN |FLEXCAN_MCR_SUPV ; + reg |=FLEXCAN_MCR_FEN |FLEXCAN_MCR_SUPV ; priv->write (reg, ®s->mcr);/* Currently we only support newer versions of this core
[email protected]
maintainerPATCH 12/168 can: flexcan: invoke flexcan_chip_freeze() to enter freeze mode
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Joakim Zhang <[email protected]>
commit c63820045e2000f05657467a08715c18c9f490d9 upstream.
Invoke flexcan_chip_freeze() to enter freeze mode, since need poll freeze mode acknowledge.
--- drivers/net/can/flexcan.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-)
drivers/net/can/flexcan.c +7 -5
@@ -1057,10 +1057,13 @@ static int flexcan_chip_start(struct netflexcan_set_bittiming (dev); +/* set freeze, halt */ + err =flexcan_chip_freeze (priv); +if (err) + goto out_chip_disable; +/* MCR * -* enable freeze -* halt now * only supervisor access * enable warning int * enable individual RX masking
@@ -1069,9 +1072,8 @@ static int flexcan_chip_start(struct net*/ reg_mcr = priv->read (®s->mcr); reg_mcr &= ~FLEXCAN_MCR_MAXMB (0xff ); - reg_mcr |=FLEXCAN_MCR_FRZ |FLEXCAN_MCR_HALT |FLEXCAN_MCR_SUPV | -FLEXCAN_MCR_WRN_EN |FLEXCAN_MCR_IRMQ |FLEXCAN_MCR_IDAM_C | -FLEXCAN_MCR_MAXMB (priv->tx_mb_idx); + reg_mcr |=FLEXCAN_MCR_SUPV |FLEXCAN_MCR_WRN_EN |FLEXCAN_MCR_IRMQ | +FLEXCAN_MCR_IDAM_C |FLEXCAN_MCR_MAXMB (priv->tx_mb_idx);/* MCR *
[email protected]
maintainerPATCH 13/168 can: tcan4x5x: tcan4x5x_init(): fix initialization - clear MRAM before entering Normal Mode
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Torin Cooper-Bennun <[email protected]>
commit 2712625200ed69c642b9abc3a403830c4643364c upstream.
This patch prevents a potentially destructive race condition. The device is fully operational on the bus after entering Normal Mode, so zeroing the MRAM after entering this mode may lead to loss of information, e.g. new received messages.
This patch fixes the problem by first initializing the MRAM, then bringing the device into Normale Mode.
--- drivers/net/can/m_can/tcan4x5x.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/can/m_can/tcan4x5x.c +3 -3
@@ -325,14 +325,14 @@ static int tcan4x5x_init(struct m_can_clif (ret)return ret; +/* Zero out the MCAN buffers */ +m_can_init_ram (cdev); + ret =regmap_update_bits (tcan4x5x->regmap,TCAN4X5X_CONFIG ,TCAN4X5X_MODE_SEL_MASK ,TCAN4X5X_MODE_NORMAL );if (ret)return ret; -/* Zero out the MCAN buffers */ -m_can_init_ram (cdev); -return ret; }
[email protected]
maintainerPATCH 14/168 tcp: add sanity tests to TCP_QUEUE_SEQ
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Eric Dumazet <[email protected]>
commit 8811f4a9836e31c14ecdf79d9f3cb7c5d463265d upstream.
Qingyu Li reported a syzkaller bug where the repro changes RCV SEQ _after_ restoring data in the receive queue.
mprotect(0x4aa000, 12288, PROT_READ) = 0mmap(0x1ffff000, 4096, PROT_NONE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x1ffff000mmap(0x20000000, 16777216, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x20000000 mmap(0x21000000, 4096, PROT_NONE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x21000000socket(AF_INET6, SOCK_STREAM, IPPROTO_IP) = 3setsockopt(3, SOL_TCP, TCP_REPAIR, [1], 4) = 0
connect(3, {sa_family=AF_INET6, sin6_port=htons(0), sin6_flowinfo=htonl(0), inet_pton(AF_INET6, "::1", &sin6_addr), sin6_scope_id=0}, 28) = 0 setsockopt(3, SOL_TCP, TCP_REPAIR_QUEUE, [1], 4) = 0
sendmsg(3, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="0x0000000000000003\0\0", iov_len=20}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 20 setsockopt(3, SOL_TCP, TCP_REPAIR, [0], 4) = 0setsockopt(3, SOL_TCP, TCP_QUEUE_SEQ, [128], 4) = 0recvfrom(3, NULL, 20, 0, NULL, NULL) = -1 ECONNRESET (Connection reset by peer)
syslog shows:
[ 111.205099] TCP recvmsg seq # bug 2: copied 80, seq 0, rcvnxt 80, fl 0 [ 111.207894] WARNING: CPU: 1 PID: 356 at net/ipv4/tcp.c:2343 tcp_recvmsg_locked+0x90e/0x29a0
This should not be allowed. TCP_QUEUE_SEQ should only be used when queues are empty.
This patch fixes this case, and the tx path as well.
--- net/ipv4/tcp.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-)
net/ipv4/tcp.c +15 -8
@@ -2957,16 +2957,23 @@ static int do_tcp_setsockopt(struct sockbreak ;case TCP_QUEUE_SEQ : -if (sk->sk_state !=TCP_CLOSE ) +if (sk->sk_state !=TCP_CLOSE ) { err = -EPERM ; -else if (tp->repair_queue ==TCP_SEND_QUEUE ) -WRITE_ONCE (tp->write_seq, val); -else if (tp->repair_queue ==TCP_RECV_QUEUE ) { -WRITE_ONCE (tp->rcv_nxt, val); -WRITE_ONCE (tp->copied_seq, val); - } -else + }else if (tp->repair_queue ==TCP_SEND_QUEUE ) { +if (!tcp_rtx_queue_empty (sk)) + err = -EPERM ; +else +WRITE_ONCE (tp->write_seq, val); + }else if (tp->repair_queue ==TCP_RECV_QUEUE ) { +if (tp->rcv_nxt != tp->copied_seq) { + err = -EPERM ; + }else { +WRITE_ONCE (tp->rcv_nxt, val); +WRITE_ONCE (tp->copied_seq, val); + } + }else { err = -EINVAL ; + }break ;case TCP_REPAIR_OPTIONS :
[email protected]
maintainerPATCH 15/168 netfilter: nf_nat: undo erroneous tcp edemux lookup
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Florian Westphal <[email protected]>
commit 03a3ca37e4c6478e3a84f04c8429dd5889e107fd upstream.
Under extremely rare conditions TCP early demux will retrieve the wrong socket.
1. local machine establishes a connection to a remote server, S, on port p.
This gives:
laddr:lport -> S:p
... both in tcp and conntrack.
2. local machine establishes a connection to host H, on port p2. 2a. TCP stack choses same laddr:lport, so we have laddr:lport -> H:p2 from TCP point of view.
2b). There is a destination NAT rewrite in place, translating
H:p2 to S:p. This results in following conntrack entries:
I) laddr:lport -> S:p (origin) S:p -> laddr:lport (reply) II) laddr:lport -> H:p2 (origin) S:p -> laddr:lport2 (reply)
NAT engine has rewritten laddr:lport to laddr:lport2 to map the reply packet to the correct origin.
When server sends SYN/ACK to laddr:lport2, the PREROUTING hook will undo-the SNAT transformation, rewriting IP header to S:p -> laddr:lport
This causes TCP early demux to associate the skb with the TCP socket of the first connection.
The INPUT hook will then reverse the DNAT transformation, rewriting the IP header to H:p2 -> laddr:lport.
Because packet ends up with the wrong socket, the new connection never completes: originator stays in SYN_SENT and conntrack entry remains in SYN_RECV until timeout, and responder retransmits SYN/ACK until it gives up.
To resolve this, orphan the skb after the input rewrite:
Because the source IP address changed, the socket must be incorrect. We can't move the DNAT undo to prerouting due to backwards compatibility, doing so will make iptables/nftables rules to no longer match the way they did.
After orphan, the packet will be handed to the next protocol layer (tcp, udp, ...) and that will repeat the socket lookup just like as if early demux was disabled.
--- net/netfilter/nf_nat_proto.c | 25 +++++++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-)
net/netfilter/nf_nat_proto.c +21 -4
@@ -646,8 +646,8 @@ nf_nat_ipv4_fn(void *priv, struct sk_buf }static unsigned int -nf_nat_ipv4_in (void *priv,struct sk_buff *skb, -const struct nf_hook_state *state) +nf_nat_ipv4_pre_routing (void *priv,struct sk_buff *skb, +const struct nf_hook_state *state) {unsigned int ret;__be32 daddr =ip_hdr (skb)->daddr;
@@ -660,6 +660,23 @@ nf_nat_ipv4_in(void *priv, struct sk_buf }static unsigned int +nf_nat_ipv4_local_in (void *priv,struct sk_buff *skb, +const struct nf_hook_state *state) +{ +__be32 saddr =ip_hdr (skb)->saddr; +struct sock *sk = skb->sk; +unsigned int ret; + + ret =nf_nat_ipv4_fn (priv, skb, state); + +if (ret ==NF_ACCEPT && sk && saddr !=ip_hdr (skb)->saddr && + !inet_sk_transparent (sk)) +skb_orphan (skb);/* TCP edemux obtained wrong socket */ + +return ret; +} + +static unsigned int nf_nat_ipv4_out (void *priv,struct sk_buff *skb,const struct nf_hook_state *state) {
@@ -736,7 +753,7 @@ nf_nat_ipv4_local_fn(void *priv, structstatic const struct nf_hook_ops nf_nat_ipv4_ops[] = {/* Before packet filtering, change destination */ { - .hook = nf_nat_ipv4_in, + .hook = nf_nat_ipv4_pre_routing, .pf =NFPROTO_IPV4 , .hooknum =NF_INET_PRE_ROUTING , .priority =NF_IP_PRI_NAT_DST ,
@@ -757,7 +774,7 @@ static const struct nf_hook_ops nf_nat_i },/* After packet filtering, change source */ { - .hook = nf_nat_ipv4_fn, + .hook = nf_nat_ipv4_local_in, .pf = NFPROTO_IPV4, .hooknum = NF_INET_LOCAL_IN, .priority = NF_IP_PRI_NAT_SRC,
[email protected]
maintainerPATCH 16/168 netfilter: x_tables: gpf inside xt_find_revision()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Vasily Averin <[email protected]>
commit 8e24edddad152b998b37a7f583175137ed2e04a5 upstream.
nested target/match_revfn() calls work with xt[NFPROTO_UNSPEC] lists without taking xt[NFPROTO_UNSPEC].mutex. This can race with module unload and cause host to crash:
general protection fault: 0000 [#1] Modules linked in: ... [last unloaded: xt_cluster] CPU: 0 PID: 542455 Comm: iptables RIP: 0010:[<ffffffff8ffbd518>] [<ffffffff8ffbd518>] strcmp+0x18/0x40 RDX: 0000000000000003 RSI: ffff9a5a5d9abe10 RDI: dead000000000111 R13: ffff9a5a5d9abe10 R14: ffff9a5a5d9abd8c R15: dead000000000100 (VvS: %R15 -- &xt_match, %RDI -- &xt_match.name, xt_cluster unregister match in xt[NFPROTO_UNSPEC].match list) Call Trace: [<ffffffff902ccf44>] match_revfn+0x54/0xc0 [<ffffffff902ccf9f>] match_revfn+0xaf/0xc0 [<ffffffff902cd01e>] xt_find_revision+0x6e/0xf0 [<ffffffffc05a5be0>] do_ipt_get_ctl+0x100/0x420 [ip_tables] [<ffffffff902cc6bf>] nf_getsockopt+0x4f/0x70 [<ffffffff902dd99e>] ip_getsockopt+0xde/0x100 [<ffffffff903039b5>] raw_getsockopt+0x25/0x50 [<ffffffff9026c5da>] sock_common_getsockopt+0x1a/0x20 [<ffffffff9026b89d>] SyS_getsockopt+0x7d/0xf0 [<ffffffff903cbf92>] system_call_fastpath+0x25/0x2a
--- net/netfilter/x_tables.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
net/netfilter/x_tables.c +4 -2
@@ -330,6 +330,7 @@ static int match_revfn(u8 af, const charconst struct xt_match *m;int have_rev =0 ; +mutex_lock (&xt[af].mutex);list_for_each_entry (m, &xt[af].match, list) {if (strcmp (m->name, name) ==0 ) {if (m->revision > *bestp)
@@ -338,6 +339,7 @@ static int match_revfn(u8 af, const char have_rev =1 ; } } +mutex_unlock (&xt[af].mutex);if (af !=NFPROTO_UNSPEC && !have_rev)return match_revfn (NFPROTO_UNSPEC , name, revision, bestp);
@@ -350,6 +352,7 @@ static int target_revfn(u8 af, const chaconst struct xt_target *t;int have_rev =0 ; +mutex_lock (&xt[af].mutex);list_for_each_entry (t, &xt[af].target, list) {if (strcmp (t->name, name) ==0 ) {if (t->revision > *bestp)
@@ -358,6 +361,7 @@ static int target_revfn(u8 af, const cha have_rev =1 ; } } +mutex_unlock (&xt[af].mutex);if (af !=NFPROTO_UNSPEC && !have_rev)return target_revfn (NFPROTO_UNSPEC , name, revision, bestp);
@@ -371,12 +375,10 @@ int xt_find_revision(u8 af, const char * {int have_rev, best =-1 ; -mutex_lock (&xt[af].mutex);if (target ==1 ) have_rev =target_revfn (af, name, revision, &best);else have_rev =match_revfn (af, name, revision, &best); -mutex_unlock (&xt[af].mutex);/* Nothing at all? Return 0 to try loading module. */ if (best ==-1 ) {
[email protected]
maintainerPATCH 17/168 selftests/bpf: No need to drop the packet when there is no geneve opt
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Hangbin Liu <[email protected]>
commit 557c223b643a35effec9654958d8edc62fd2603a upstream.
In bpf geneve tunnel test we set geneve option on tx side. On rx side we only call bpf_skb_get_tunnel_opt(). Since commit 9c2e14b48119 ("ip_tunnels:
Set tunnel option flag when tunnel metadata is present") geneve_rx() will not add TUNNEL_GENEVE_OPT flag if there is no geneve option, which cause bpf_skb_get_tunnel_opt() return ENOENT and _geneve_get_tunnel() in test_tunnel_kern.c drop the packet.
As it should be valid that bpf_skb_get_tunnel_opt() return error when there is not tunnel option, there is no need to drop the packet and break all geneve rx traffic. Just set opt_class to 0 in this test and keep returning TC_ACT_OK.
--- tools/testing/selftests/bpf/progs/test_tunnel_kern.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
tools/testing/selftests/bpf/progs/test_tunnel_kern.c +2 -4
@@ -446,10 +446,8 @@ int _geneve_get_tunnel(struct __sk_buff } ret =bpf_skb_get_tunnel_opt (skb, &gopt,sizeof (gopt)); -if (ret <0 ) { -ERROR (ret); -return TC_ACT_SHOT ; - } +if (ret <0 ) + gopt.opt_class =0 ;bpf_trace_printk (fmt,sizeof (fmt), key.tunnel_id, key.remote_ipv4, gopt.opt_class);
[email protected]
maintainerPATCH 18/168 selftests/bpf: Mask bpf_csum_diff() return value to 16 bits in test_verifier
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Yauheni Kaliuta <[email protected]>
commit 6185266c5a853bb0f2a459e3ff594546f277609b upstream.
The verifier test labelled "valid read map access into a read-only array 2" calls the bpf_csum_diff() helper and checks its return value. However, architecture implementations of csum_partial() (which is what the helper uses) differ in whether they fold the return value to 16 bit or not. For example, x86 version has ...
if (unlikely(odd)) {
result = from32to16(result);
result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
}... while generic lib/checksum.c does:
result = from32to16(result); if (odd) result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
This makes the helper return different values on different architectures, breaking the test on non-x86. To fix this, add an additional instruction to always mask the return value to 16 bits, and update the expected return value accordingly.
--- tools/testing/selftests/bpf/verifier/array_access.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
tools/testing/selftests/bpf/verifier/array_access.c +2 -1
@@ -250,12 +250,13 @@BPF_MOV64_IMM (BPF_REG_5 ,0 ),BPF_RAW_INSN (BPF_JMP |BPF_CALL ,0 ,0 ,0 , BPF_FUNC_csum_diff), +BPF_ALU64_IMM (BPF_AND ,BPF_REG_0 ,0xffff ), BPF_EXIT_INSN(), }, .prog_type =BPF_PROG_TYPE_SCHED_CLS , .fixup_map_array_ro = { 3 }, .result = ACCEPT, - .retval = -29, + .retval = 65507, }, { "invalid write map access into a read-only array 1",
[email protected]
maintainerPATCH 19/168 samples, bpf: Add missing munmap in xdpsock
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Maciej Fijalkowski <[email protected]>
commit 6bc6699881012b5bd5d49fa861a69a37fc01b49c upstream.
We mmap the umem region, but we never munmap it.
Add the missing call at the end of the cleanup.
--- samples/bpf/xdpsock_user.c | 2 ++ 1 file changed, 2 insertions(+)
samples/bpf/xdpsock_user.c +2 -0
@@ -783,5 +783,7 @@ int main(int argc, char **argv)else l2fwd_all (); +munmap (bufs,NUM_FRAMES * opt_xsk_frame_size); +return 0 ; }
[email protected]
maintainerPATCH 20/168 ibmvnic: always store valid MAC address
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Jiri Wiesner <[email protected]>
commit 67eb211487f0c993d9f402d1c196ef159fd6a3b5 upstream.
The last change to ibmvnic_set_mac(), 8fc3672a8ad3, meant to prevent users from setting an invalid MAC address on an ibmvnic interface that has not been brought up yet. The change also prevented the requested MAC address from being stored by the adapter object for an ibmvnic interface when the state of the ibmvnic interface is VNIC_PROBED - that is after probing has finished but before the ibmvnic interface is brought up. The MAC address stored by the adapter object is used and sent to the hypervisor for checking when an ibmvnic interface is brought up.
The ibmvnic driver ignoring the requested MAC address when in VNIC_PROBED state caused LACP bonds (bonds in 802.3ad mode) with more than one slave to malfunction. The bonding code must be able to change the MAC address of its slaves before they are brought up during enslaving. The inability of kernels with 8fc3672a8ad3 to set the MAC addresses of bonding slaves is observable in the output of "ip address show". The MAC addresses of the slaves are the same as the MAC address of the bond on a working system whereas the slaves retain their original MAC addresses on a system with a malfunctioning LACP bond.
--- drivers/net/ethernet/ibm/ibmvnic.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/net/ethernet/ibm/ibmvnic.c +2 -3
@@ -1753,10 +1753,9 @@ static int ibmvnic_set_mac(struct net_deif (!is_valid_ether_addr (addr->sa_data))return -EADDRNOTAVAIL ; -if (adapter->state !=VNIC_PROBED ) { -ether_addr_copy (adapter->mac_addr, addr->sa_data); +ether_addr_copy (adapter->mac_addr, addr->sa_data); +if (adapter->state !=VNIC_PROBED ) rc =__ibmvnic_set_mac (netdev, addr->sa_data); - }return rc; }
[email protected]
maintainerPATCH 21/168 mt76: dma: do not report truncated frames to mac80211
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Lorenzo Bianconi <[email protected]>
commit d0bd52c591a1070c54dc428e926660eb4f981099 upstream.
Commit b102f0c522cf6 ("mt76: fix array overflow on receiving too many fragments for a packet") fixes a possible OOB access but it introduces a memory leak since the pending frame is not released to page_frag_cache if the frag array of skb_shared_info is full. Commit 93a1d4791c10 ("mt76: dma: fix a possible memory leak in mt76_add_fragment()") fixes the issue but does not free the truncated skb that is forwarded to mac80211 layer. Fix the leftover issue discarding even truncated skbs.
--- drivers/net/wireless/mediatek/mt76/dma.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/net/wireless/mediatek/mt76/dma.c +7 -4
@@ -454,13 +454,13 @@ mt76_add_fragment(struct mt76_dev *dev, {struct sk_buff *skb = q->rx_head;struct skb_shared_info *shinfo =skb_shinfo (skb); +int nr_frags = shinfo->nr_frags; -if (shinfo->nr_frags <ARRAY_SIZE (shinfo->frags)) { +if (nr_frags <ARRAY_SIZE (shinfo->frags)) {struct page *page =virt_to_head_page (data);int offset = data -page_address (page) + q->buf_offset; -skb_add_rx_frag (skb, shinfo->nr_frags, page, offset, len, - q->buf_size); +skb_add_rx_frag (skb, nr_frags, page, offset, len, q->buf_size); }else {skb_free_frag (data); }
@@ -469,7 +469,10 @@ mt76_add_fragment(struct mt76_dev *dev,return ; q->rx_head =NULL ; - dev->drv->rx_skb (dev, q - dev->q_rx, skb); +if (nr_frags <ARRAY_SIZE (shinfo->frags)) + dev->drv->rx_skb (dev, q - dev->q_rx, skb); +else +dev_kfree_skb (skb); } static int
[email protected]
maintainerPATCH 22/168 powerpc/603: Fix protection of user pages mapped with PROT_NONE
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Christophe Leroy <[email protected]>
commit c119565a15a628efdfa51352f9f6c5186e506a1c upstream.
On book3s/32, page protection is defined by the PP bits in the PTE which provide the following protection depending on the access keys defined in the matching segment register:
- PP 00 means RW with key 0 and N/A with key 1.
- PP 01 means RW with key 0 and RO with key 1.
- PP 10 means RW with both key 0 and key 1.
- PP 11 means RO with both key 0 and key 1.
Since the implementation of kernel userspace access protection, PP bits have been set as follows:
- PP00 for pages without _PAGE_USER
- PP01 for pages with _PAGE_USER and _PAGE_RW
- PP11 for pages with _PAGE_USER and without _PAGE_RW
For kernelspace segments, kernel accesses are performed with key 0 and user accesses are performed with key 1. As PP00 is used for non _PAGE_USER pages, user can't access kernel pages not flagged _PAGE_USER while kernel can.
For userspace segments, both kernel and user accesses are performed with key 0, therefore pages not flagged _PAGE_USER are still accessible to the user.
This shouldn't be an issue, because userspace is expected to be accessible to the user. But unlike most other architectures, powerpc implements PROT_NONE protection by removing _PAGE_USER flag instead of flagging the page as not valid. This means that pages in userspace that are not flagged _PAGE_USER shall remain inaccessible.
To get the expected behaviour, just mimic other architectures in the TLB miss handler by checking _PAGE_USER permission on userspace accesses as if it was the _PAGE_PRESENT bit.
Note that this problem only is only for 603 cores. The 604+ have an hash table, and hash_page() function already implement the verification of _PAGE_USER permission on userspace pages.
--- arch/powerpc/kernel/head_32.S | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
arch/powerpc/kernel/head_32.S +6 -3
@@ -418,10 +418,11 @@ InstructionTLBMiss: cmplw 0,r1,r3 #endif mfspr r2, SPRN_SPRG_PGDIR - li r1,_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_EXEC + li r1,_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_EXEC | _PAGE_USER #if defined(CONFIG_MODULES) || defined(CONFIG_DEBUG_PAGEALLOC) bge- 112f lis r2, (swapper_pg_dir - PAGE_OFFSET)@ha /* if kernel address, use */ + li r1,_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_EXEC addi r2, r2, (swapper_pg_dir - PAGE_OFFSET)@l /* kernel page table */ #endif 112: rlwimi r2,r3,12,20,29 /* insert top 10 bits of address */
@@ -480,9 +481,10 @@ DataLoadTLBMiss: lis r1,PAGE_OFFSET@h /* check if kernel address */ cmplw 0,r1,r3 mfspr r2, SPRN_SPRG_PGDIR - li r1, _PAGE_PRESENT | _PAGE_ACCESSED + li r1, _PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_USER bge- 112f lis r2, (swapper_pg_dir - PAGE_OFFSET)@ha /* if kernel address, use */ + li r1, _PAGE_PRESENT | _PAGE_ACCESSED addi r2, r2, (swapper_pg_dir - PAGE_OFFSET)@l /* kernel page table */ 112: rlwimi r2,r3,12,20,29 /* insert top 10 bits of address */ lwz r2,0(r2) /* get pmd entry */
@@ -556,9 +558,10 @@ DataStoreTLBMiss: lis r1,PAGE_OFFSET@h /* check if kernel address */ cmplw 0,r1,r3 mfspr r2, SPRN_SPRG_PGDIR - li r1, _PAGE_RW | _PAGE_DIRTY | _PAGE_PRESENT | _PAGE_ACCESSED + li r1, _PAGE_RW | _PAGE_DIRTY | _PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_USER bge- 112f lis r2, (swapper_pg_dir - PAGE_OFFSET)@ha /* if kernel address, use */ + li r1, _PAGE_RW | _PAGE_DIRTY | _PAGE_PRESENT | _PAGE_ACCESSED addi r2, r2, (swapper_pg_dir - PAGE_OFFSET)@l /* kernel page table */ 112: rlwimi r2,r3,12,20,29 /* insert top 10 bits of address */ lwz r2,0(r2) /* get pmd entry */
[email protected]
maintainerPATCH 23/168 mount: fix mounting of detached mounts onto targets that reside on shared mounts
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Christian Brauner <[email protected]>
commit ee2e3f50629f17b0752b55b2566c15ce8dafb557 upstream.
Creating a series of detached mounts, attaching them to the filesystem, and unmounting them can be used to trigger an integer overflow in ns->mounts causing the kernel to block any new mounts in count_mounts() and returning ENOSPC because it falsely assumes that the maximum number of mounts in the mount namespace has been reached, i.e. it thinks it can't fit the new mounts into the mount namespace anymore.
Depending on the number of mounts in your system, this can be reproduced on any kernel that supportes open_tree() and move_mount() by compiling and running the following program:
/* SPDX-License-Identifier: LGPL-2.1+ */
#define _GNU_SOURCE
#include <errno.h>
#include <fcntl.h>
#include <getopt.h>
#include <limits.h>
#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/mount.h>
#include <sys/stat.h>
#include <sys/syscall.h>
#include <sys/types.h>
#include <unistd.h>
/* open_tree() */
#ifndef OPEN_TREE_CLONE
#define OPEN_TREE_CLONE 1
#endif
#ifndef OPEN_TREE_CLOEXEC
#define OPEN_TREE_CLOEXEC O_CLOEXEC
#endif
#ifndef __NR_open_tree
#if defined __alpha__
#define __NR_open_tree 538
#elif defined _MIPS_SIM
#if _MIPS_SIM == _MIPS_SIM_ABI32 /* o32 */
#define __NR_open_tree 4428
#endif
#if _MIPS_SIM == _MIPS_SIM_NABI32 /* n32 */
#define __NR_open_tree 6428
#endif
#if _MIPS_SIM == _MIPS_SIM_ABI64 /* n64 */
#define __NR_open_tree 5428
#endif
#elif defined __ia64__
#define __NR_open_tree (428 + 1024)
#else
#define __NR_open_tree 428
#endif
#endif
/* move_mount() */
#ifndef MOVE_MOUNT_F_EMPTY_PATH
#define MOVE_MOUNT_F_EMPTY_PATH 0x00000004 /* Empty from path permitted */
#endif
#ifndef __NR_move_mount
#if defined __alpha__
#define __NR_move_mount 539
#elif defined _MIPS_SIM
#if _MIPS_SIM == _MIPS_SIM_ABI32 /* o32 */
#define __NR_move_mount 4429
#endif
#if _MIPS_SIM == _MIPS_SIM_NABI32 /* n32 */
#define __NR_move_mount 6429
#endif
#if _MIPS_SIM == _MIPS_SIM_ABI64 /* n64 */
#define __NR_move_mount 5429
#endif
#elif defined __ia64__
#define __NR_move_mount (428 + 1024)
#else
#define __NR_move_mount 429
#endif
#endif
static inline int sys_open_tree(int dfd, const char *filename, unsigned int flags)
{
return syscall(__NR_open_tree, dfd, filename, flags);
}
static inline int sys_move_mount(int from_dfd, const char *from_pathname, int to_dfd,
const char *to_pathname, unsigned int flags)
{
return syscall(__NR_move_mount, from_dfd, from_pathname, to_dfd, to_pathname, flags);
}
static bool is_shared_mountpoint(const char *path)
{
bool shared = false;
FILE *f = NULL;
char *line = NULL;
int i;
size_t len = 0; f = fopen("/proc/self/mountinfo", "re");
if (!f)
return 0;
while (getline(&line, &len, f) > 0) {
char *slider1, *slider2;
for (slider1 = line, i = 0; slider1 && i < 4; i++)
slider1 = strchr(slider1 + 1, ' ');
if (!slider1)
continue;
slider2 = strchr(slider1 + 1, ' ');
if (!slider2)
continue;
*slider2 = '\0';
if (strcmp(slider1 + 1, path) == 0) {
/* This is the path. Is it shared? */
slider1 = strchr(slider2 + 1, ' ');
if (slider1 && strstr(slider1, "shared:")) {
shared = true;
break;
}
}
}
fclose(f);
free(line);
return shared;}
static void usage(void)
{
const char *text = "mount-new [--recursive] <base-dir>\n";
fprintf(stderr, "%s", text);
_exit(EXIT_SUCCESS);
}
#define exit_usage(format, ...) \
({ \
fprintf(stderr, format "\n", ##__VA_ARGS__); \
usage(); \
})
#define exit_log(format, ...) \
({ \
fprintf(stderr, format "\n", ##__VA_ARGS__); \
exit(EXIT_FAILURE); \
})static const struct option longopts[] = {
{"help", no_argument, 0, 'a'},
{ NULL, no_argument, 0, 0 },};
int main(int argc, char *argv[])
{
int exit_code = EXIT_SUCCESS, index = 0;
int dfd, fd_tree, new_argc, ret;
char *base_dir;
char *const *new_argv;
char target[PATH_MAX]; while ((ret = getopt_long_only(argc, argv, "", longopts, &index)) != -1) {
switch (ret) {
case 'a':
/* fallthrough */
default:
usage();
}
}
new_argv = &argv[optind];
new_argc = argc - optind;
if (new_argc < 1)
exit_usage("Missing base directory\n");
base_dir = new_argv[0];
if (*base_dir != '/')
exit_log("Please specify an absolute path");
/* Ensure that target is a shared mountpoint. */
if (!is_shared_mountpoint(base_dir))
exit_log("Please ensure that \"%s\" is a shared mountpoint", base_dir);
dfd = open(base_dir, O_RDONLY | O_DIRECTORY | O_CLOEXEC);
if (dfd < 0)
exit_log("%m - Failed to open base directory \"%s\"", base_dir);
ret = mkdirat(dfd, "detached-move-mount", 0755);
if (ret < 0)
exit_log("%m - Failed to create required temporary directories");
ret = snprintf(target, sizeof(target), "%s/detached-move-mount", base_dir);
if (ret < 0 || (size_t)ret >= sizeof(target))
exit_log("%m - Failed to assemble target path");
/*
* Having a mount table with 10000 mounts is already quite excessive
* and shoult account even for weird test systems.
*/
for (size_t i = 0; i < 10000; i++) {
fd_tree = sys_open_tree(dfd, "detached-move-mount",
OPEN_TREE_CLONE |
OPEN_TREE_CLOEXEC |
AT_EMPTY_PATH);
if (fd_tree < 0) {
fprintf(stderr, "%m - Failed to open %d(detached-move-mount)", dfd);
exit_code = EXIT_FAILURE;
break;
}
ret = sys_move_mount(fd_tree, "", dfd, "detached-move-mount", MOVE_MOUNT_F_EMPTY_PATH);
if (ret < 0) {
if (errno == ENOSPC)
fprintf(stderr, "%m - Buggy mount counting");
else
fprintf(stderr, "%m - Failed to attach mount to %d(detached-move-mount)", dfd);
exit_code = EXIT_FAILURE;
break;
}
close(fd_tree);
ret = umount2(target, MNT_DETACH);
if (ret < 0) {
fprintf(stderr, "%m - Failed to unmount %s", target);
exit_code = EXIT_FAILURE;
break;
}
}
(void)unlinkat(dfd, "detached-move-mount", AT_REMOVEDIR);
close(dfd);
exit(exit_code);}
and wait for the kernel to refuse any new mounts by returning ENOSPC. How many iterations are needed depends on the number of mounts in your system. Assuming you have something like 50 mounts on a standard system it should be almost instantaneous.
The root cause of this is that detached mounts aren't handled correctly when source and target mount are identical and reside on a shared mount causing a broken mount tree where the detached source itself is propagated which propagation prevents for regular bind-mounts and new mounts. This ultimately leads to a miscalculation of the number of mounts in the mount namespace.
Detached mounts created viaopen_tree(fd, path, OPEN_TREE_CLONE)
are essentially like an unattached new mount, or an unattached bind-mount. They can then later on be attached to the filesystem via move_mount() which calls into attach_recursive_mount(). Part of attaching it to the filesystem is making sure that mounts get correctly propagated in case the destination mountpoint is MS_SHARED, i.e. is a shared mountpoint. This is done by calling into propagate_mnt() which walks the list of peers calling propagate_one() on each mount in this list making sure it receives the propagation event.
The propagate_one() functions thereby skips both new mounts and bind mounts to not propagate them "into themselves". Both are identified by checking whether the mount is already attached to any mount namespace in mnt->mnt_ns. The is what the IS_MNT_NEW() helper is responsible for.
However, detached mounts have an anonymous mount namespace attached to them stashed in mnt->mnt_ns which means that IS_MNT_NEW() doesn't realize they need to be skipped causing the mount to propagate "into itself" breaking the mount table and causing a disconnect between the number of mounts recorded as being beneath or reachable from the target mountpoint and the number of mounts actually recorded/counted in ns->mounts ultimately causing an overflow which in turn prevents any new mounts via the ENOSPC issue.
So teach propagation to handle detached mounts by making it aware of them. I've been tracking this issue down for the last couple of days and then verifying that the fix is correct by
unmounting everything in my current mount table leaving only /proc and /sys mounted and running the reproducer above overnight verifying the number of mounts counted in ns->mounts. With this fix the counts are correct and the ENOSPC issue can't be reproduced.
This change will only have an effect on mounts created with the new mount API since detached mounts cannot be created with the old mount API so regressions are extremely unlikely.
--- fs/pnode.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/pnode.h +1 -1
@@ -12,7 +12,7 @@#define IS_MNT_SHARED (m) ((m)->mnt.mnt_flags & MNT_SHARED)#define IS_MNT_SLAVE (m) ((m)->mnt_master) -#define IS_MNT_NEW (m) (!(m)->mnt_ns) +#define IS_MNT_NEW (m) (!(m)->mnt_ns || is_anon_ns((m)->mnt_ns))#define CLEAR_MNT_SHARED (m) ((m)->mnt.mnt_flags &= ~MNT_SHARED)#define IS_MNT_UNBINDABLE (m) ((m)->mnt.mnt_flags & MNT_UNBINDABLE)#define IS_MNT_MARKED (m) ((m)->mnt.mnt_flags & MNT_MARKED)
[email protected]
maintainerPATCH 24/168 cifs: return proper error code in statfs(2)
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Paulo Alcantara <[email protected]>
commit 14302ee3301b3a77b331cc14efb95bf7184c73cc upstream.
In cifs_statfs(), if server->ops->queryfs is not NULL, then we should use its return value rather than always returning 0. Instead, use rc variable as it is properly set to 0 in case there is no server->ops->queryfs.
--- fs/cifs/cifsfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/cifs/cifsfs.c +1 -1
@@ -278,7 +278,7 @@ cifs_statfs(struct dentry *dentry, struc rc = server->ops->queryfs (xid, tcon, buf);free_xid (xid); -return 0 ; +return rc; }static long cifs_fallocate (struct file *file,int mode,loff_t off,loff_t len)
[email protected]
maintainerPATCH 25/168 Revert "mm, slub: consider rest of partial list if acquire_slab() fails"
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Linus Torvalds <[email protected]>
commit 9b1ea29bc0d7b94d420f96a0f4121403efc3dd85 upstream.
This reverts commit 8ff60eb052eeba95cfb3efe16b08c9199f8121cf.
The kernel test robot reports a huge performance regression due to the commit, and the reason seems fairly straightforward: when there is contention on the page list (which is what causes acquire_slab() to fail), we do _not_ want to just loop and try again, because that will transfer the contention to the 'n->list_lock' spinlock we hold, and just make things even worse.
This is admittedly likely a problem only on big machines - the kernel test robot report comes from a 96-thread dual socket Intel Xeon Gold 6252 setup, but the regression there really is quite noticeable:
-47.9% regression of stress-ng.rawpkt.ops_per_sec
and the commit that was marked as being fixed (7ced37197196: "slub:Acquire_slab() avoid loop") actually did the loop exit early very intentionally (the hint being that "avoid loop" part of that commit message), exactly to avoid this issue.
The correct thing to do may be to pick some kind of reasonable middle
ground: instead of breaking out of the loop on the very first sign of contention, or trying over and over and over again, the right thing may be to re-try _once_, and then give up on the second failure (or pick your favorite value for "once"..).
--- mm/slub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
mm/slub.c +1 -1
@@ -1887,7 +1887,7 @@ static void *get_partial_node(struct kme t =acquire_slab (s, n, page, object ==NULL , &objects);if (!t) -continue ;/* cmpxchg raced */ +break ; available += objects;if (!object) {
[email protected]
maintainerPATCH 26/168 net: enetc: dont overwrite the RSS indirection table when initializing
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Vladimir Oltean <[email protected]>
commit c646d10dda2dcde82c6ce5a474522621ab2b8b19 upstream.
After the blamed patch, all RX traffic gets hashed to CPU 0 because the hashing indirection table set up in:
enetc_pf_probe
-> enetc_alloc_si_resources
-> enetc_configure_si
-> enetc_setup_default_rss_table
is overwritten later in:
enetc_pf_probe
-> enetc_init_port_rss_memory
which zero-initializes the entire port RSS table in order to avoid ECC errors.
The trouble really is that enetc_init_port_rss_memory really neads enetc_alloc_si_resources to be called, because it depends upon enetc_alloc_cbdr and enetc_setup_cbdr. But that whole enetc_configure_si thing could have been better thought out, it has nothing to do in a function called "alloc_si_resources", especially since its counterpart, "free_si_resources", does nothing to unwind the configuration of the SI.
The point is, we need to pull out enetc_configure_si out of enetc_alloc_resources, and move it after enetc_init_port_rss_memory. This allows us to set up the default RSS indirection table after initializing the memory.
--- drivers/net/ethernet/freescale/enetc/enetc.c | 11 +++-------- drivers/net/ethernet/freescale/enetc/enetc.h | 1 + drivers/net/ethernet/freescale/enetc/enetc_pf.c | 7 +++++++ drivers/net/ethernet/freescale/enetc/enetc_vf.c | 7 +++++++ 4 files changed, 18 insertions(+), 8 deletions(-)
drivers/net/ethernet/freescale/enetc/enetc.c +3 -8
@@ -1016,13 +1016,12 @@ static int enetc_setup_default_rss_tablereturn 0 ; } -static int enetc_configure_si (struct enetc_ndev_priv *priv) +int enetc_configure_si (struct enetc_ndev_priv *priv) {struct enetc_si *si = priv->si;struct enetc_hw *hw = &si->hw;int err; -enetc_setup_cbdr (hw, &si->cbd_ring);/* set SI cache attributes */ enetc_wr (hw,ENETC_SICAR0 ,ENETC_SICAR_RD_COHERENT |ENETC_SICAR_WR_COHERENT );
@@ -1068,6 +1067,8 @@ int enetc_alloc_si_resources(struct enetif (err)return err; +enetc_setup_cbdr (&si->hw, &si->cbd_ring); + priv->cls_rules =kcalloc (si->num_fs_entries,sizeof (*priv->cls_rules),GFP_KERNEL );if (!priv->cls_rules) {
@@ -1075,14 +1076,8 @@ int enetc_alloc_si_resources(struct enet goto err_alloc_cls; } - err =enetc_configure_si (priv); -if (err) - goto err_config_si; -return 0 ; -err_config_si: -kfree (priv->cls_rules); err_alloc_cls:enetc_clear_cbdr (&si->hw);enetc_free_cbdr (priv->dev, &si->cbd_ring);
drivers/net/ethernet/freescale/enetc/enetc.h +1 -0
@@ -221,6 +221,7 @@ void enetc_get_si_caps(struct enetc_si *void enetc_init_si_rings_params (struct enetc_ndev_priv *priv);int enetc_alloc_si_resources (struct enetc_ndev_priv *priv);void enetc_free_si_resources (struct enetc_ndev_priv *priv); +int enetc_configure_si (struct enetc_ndev_priv *priv);int enetc_open (struct net_device *ndev);int enetc_close (struct net_device *ndev);
drivers/net/ethernet/freescale/enetc/enetc_pf.c +7 -0
@@ -920,6 +920,12 @@ static int enetc_pf_probe(struct pci_dev goto err_init_port_rss; } + err =enetc_configure_si (priv); +if (err) { +dev_err (&pdev->dev,"Failed to configure SI\n" ); + goto err_config_si; + } + err =enetc_alloc_msix (priv);if (err) {dev_err (&pdev->dev,"MSIX alloc failed\n" );
@@ -945,6 +951,7 @@ err_reg_netdev:enetc_mdio_remove (pf);enetc_of_put_phy (priv);enetc_free_msix (priv); +err_config_si: err_init_port_rss: err_init_port_rfs: err_alloc_msix:
drivers/net/ethernet/freescale/enetc/enetc_vf.c +7 -0
@@ -189,6 +189,12 @@ static int enetc_vf_probe(struct pci_dev goto err_alloc_si_res; } + err =enetc_configure_si (priv); +if (err) { +dev_err (&pdev->dev,"Failed to configure SI\n" ); + goto err_config_si; + } + err =enetc_alloc_msix (priv);if (err) {dev_err (&pdev->dev,"MSIX alloc failed\n" );
@@ -208,6 +214,7 @@ static int enetc_vf_probe(struct pci_dev err_reg_netdev:enetc_free_msix (priv); +err_config_si: err_alloc_msix:enetc_free_si_resources (priv); err_alloc_si_res:
[email protected]
maintainerPATCH 27/168 net/mlx4_en: update moderation when config reset
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Kevin(Yudong) Yang <[email protected]>
commit 00ff801bb8ce6711e919af4530b6ffa14a22390a upstream.
This patch fixes a bug that the moderation config will not be applied when calling mlx4_en_reset_config. For example, when turning on rx timestamping, mlx4_en_reset_config() will be called, causing the NIC to forget previous moderation config.
This fix is in phase with a previous fix:
commit 79c54b6bbf06 ("net/mlx4_en: Fix TX moderation info loss after set_ringparam is called")
Tested: Before this patch, on a host with NIC using mlx4, run netserver and stream TCP to the host at full utilization.
$ sar -I SUM 1
INTR intr/s
14:03:56 sum 48758.00After rx hwtstamp is enabled:
$ sar -I SUM 1 14:10:38 sum 317771.00 We see the moderation is not working properly and issued 7x more interrupts.
After the patch, and turned on rx hwtstamp, the rate of interrupts is as expected:
$ sar -I SUM 1 14:52:11 sum 49332.00
--- drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 2 +- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 2 ++ drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 1 + 3 files changed, 4 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +1 -1
@@ -47,7 +47,7 @@#define EN_ETHTOOL_SHORT_MASK cpu_to_be16(0xffff)#define EN_ETHTOOL_WORD_MASK cpu_to_be32(0xffffffff) -static int mlx4_en_moderation_update (struct mlx4_en_priv *priv) +int mlx4_en_moderation_update (struct mlx4_en_priv *priv) {int i, t;int err =0 ;
drivers/net/ethernet/mellanox/mlx4/en_netdev.c +2 -0
@@ -3657,6 +3657,8 @@ int mlx4_en_reset_config(struct net_devien_err (priv,"Failed starting port\n" ); } +if (!err) + err =mlx4_en_moderation_update (priv); out:mutex_unlock (&mdev->state_lock);kfree (tmp);
drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +1 -0
@@ -797,6 +797,7 @@ void mlx4_en_ptp_overflow_check(struct m#define DEV_FEATURE_CHANGED (dev, new_features, feature) \ ((dev->features & feature) ^ (new_features & feature)) +int mlx4_en_moderation_update (struct mlx4_en_priv *priv);int mlx4_en_reset_config (struct net_device *dev,struct hwtstamp_config ts_config,netdev_features_t new_features);
[email protected]
maintainerPATCH 28/168 net: stmmac: fix incorrect DMA channel intr enable setting of EQoS v4.10
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ong Boon Leong <[email protected]>
commit 879c348c35bb5fb758dd881d8a97409c1862dae8 upstream.
We introduce dwmac410_dma_init_channel() here for both EQoS v4.10 and above which use different DMA_CH(n)_Interrupt_Enable bit definitions for NIE and AIE.
--- drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c +18 -1
@@ -116,6 +116,23 @@ static void dwmac4_dma_init_channel(void ioaddr + DMA_CHAN_INTR_ENA(chan)); } +static void dwmac410_dma_init_channel (void __iomem *ioaddr, +struct stmmac_dma_cfg *dma_cfg,u32 chan) +{ +u32 value; + +/* common channel control register config */ + value =readl (ioaddr +DMA_CHAN_CONTROL (chan)); +if (dma_cfg->pblx8) + value = value |DMA_BUS_MODE_PBL ; + +writel (value, ioaddr +DMA_CHAN_CONTROL (chan)); + +/* Mask interrupts by writing to CSR7 */ +writel (DMA_CHAN_INTR_DEFAULT_MASK_4_10 , + ioaddr +DMA_CHAN_INTR_ENA (chan)); +} +static void dwmac4_dma_init (void __iomem *ioaddr,struct stmmac_dma_cfg *dma_cfg,int atds) {
@@ -462,7 +479,7 @@ const struct stmmac_dma_ops dwmac4_dma_oconst struct stmmac_dma_ops dwmac410_dma_ops = { .reset = dwmac4_dma_reset, .init = dwmac4_dma_init, - .init_chan = dwmac4_dma_init_channel, + .init_chan = dwmac410_dma_init_channel, .init_rx_chan = dwmac4_dma_init_rx_chan, .init_tx_chan = dwmac4_dma_init_tx_chan, .axi = dwmac4_dma_axi,
[email protected]
maintainerPATCH 29/168 nexthop: Do not flush blackhole nexthops when loopback goes down
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ido Schimmel <[email protected]>
commit 76c03bf8e2624076b88d93542d78e22d5345c88e upstream.
As far as user space is concerned, blackhole nexthops do not have a nexthop device and therefore should not be affected by the administrative or carrier state of any netdev.
However, when the loopback netdev goes down all the blackhole nexthops are flushed. This happens because internally the kernel associates blackhole nexthops with the loopback netdev.
This behavior is both confusing to those not familiar with kernel internals and also diverges from the legacy API where blackhole IPv4 routes are not flushed when the loopback netdev goes down:
# ip route add blackhole 198.51.100.0/24 # ip link set dev lo down # ip route show 198.51.100.0/24 blackhole 198.51.100.0/24
Blackhole IPv6 routes are flushed, but at least user space knows that they are associated with the loopback netdev:
# ip -6 route show 2001:db8:1::/64 blackhole 2001:db8:1::/64 dev lo metric 1024 pref medium
Fix this by only flushing blackhole nexthops when the loopback netdev is unregistered.
--- net/ipv4/nexthop.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
net/ipv4/nexthop.c +7 -3
@@ -1065,7 +1065,7 @@ out:/* rtnl */ /* remove all nexthops tied to a device being deleted */ -static void nexthop_flush_dev (struct net_device *dev) +static void nexthop_flush_dev (struct net_device *dev,unsigned long event) {unsigned int hash =nh_dev_hashfn (dev->ifindex);struct net *net =dev_net (dev);
@@ -1077,6 +1077,10 @@ static void nexthop_flush_dev(struct netif (nhi->fib_nhc.nhc_dev != dev)continue ; +if (nhi->reject_nh && + (event ==NETDEV_DOWN || event ==NETDEV_CHANGE )) +continue ; +remove_nexthop (net, nhi->nh_parent,NULL ); } }
@@ -1794,11 +1798,11 @@ static int nh_netdev_event(struct notifiswitch (event) {case NETDEV_DOWN :case NETDEV_UNREGISTER : -nexthop_flush_dev (dev); +nexthop_flush_dev (dev, event);break ;case NETDEV_CHANGE :if (!(dev_get_flags (dev) & (IFF_RUNNING |IFF_LOWER_UP ))) -nexthop_flush_dev (dev); +nexthop_flush_dev (dev, event);break ;case NETDEV_CHANGEMTU : info_ext = ptr;
[email protected]
maintainerPATCH 30/168 net: sched: avoid duplicates in classes dump
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Maximilian Heyne <[email protected]>
commit bfc2560563586372212b0a8aeca7428975fa91fe upstream.
This is a follow up of commit ea3274695353 ("net: sched: avoid duplicates in qdisc dump") which has fixed the issue only for the qdisc dump.
The duplicate printing also occurs when dumping the classes via
tc class show dev eth0
--- net/sched/sch_api.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
net/sched/sch_api.c +4 -4
@@ -2157,7 +2157,7 @@ static int tc_dump_tclass_qdisc(struct Qstatic int tc_dump_tclass_root (struct Qdisc *root,struct sk_buff *skb,struct tcmsg *tcm,struct netlink_callback *cb, -int *t_p,int s_t) +int *t_p,int s_t,bool recur) {struct Qdisc *q;int b;
@@ -2168,7 +2168,7 @@ static int tc_dump_tclass_root(struct Qdif (tc_dump_tclass_qdisc (root, skb, tcm, cb, t_p, s_t) <0 )return -1 ; -if (!qdisc_dev (root)) +if (!qdisc_dev (root) || !recur)return 0 ;if (tcm->tcm_parent) {
@@ -2203,13 +2203,13 @@ static int tc_dump_tclass(struct sk_buff s_t = cb->args[0 ]; t =0 ; -if (tc_dump_tclass_root (dev->qdisc, skb, tcm, cb, &t, s_t) <0 ) +if (tc_dump_tclass_root (dev->qdisc, skb, tcm, cb, &t, s_t, true) <0 ) goto done; dev_queue =dev_ingress_queue (dev);if (dev_queue &&tc_dump_tclass_root (dev_queue->qdisc_sleeping, skb, tcm, cb, - &t, s_t) <0 ) + &t, s_t, false) <0 ) goto done; done:
[email protected]
maintainerPATCH 31/168 net: usb: qmi_wwan: allow qmimux add/del with master up
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Daniele Palmas <[email protected]>
commit 6c59cff38e66584ae3ac6c2f0cbd8d039c710ba7 upstream.
There's no reason for preventing the creation and removal of qmimux network interfaces when the underlying interface is up.
This makes qmi_wwan mux implementation more similar to the rmnet one, simplifying userspace management of the same logical interfaces.
--- drivers/net/usb/qmi_wwan.c | 14 -------------- 1 file changed, 14 deletions(-)
drivers/net/usb/qmi_wwan.c +0 -14
@@ -441,13 +441,6 @@ static ssize_t add_mux_store(struct devi goto err; } -/* we don't want to modify a running netdev */ -if (netif_running (dev->net)) { -netdev_err (dev->net,"Cannot change a running device\n" ); - ret = -EBUSY ; - goto err; - } - ret =qmimux_register_device (dev->net, mux_id);if (!ret) { info->flags |=QMI_WWAN_FLAG_MUX ;
@@ -477,13 +470,6 @@ static ssize_t del_mux_store(struct deviif (!rtnl_trylock ())return restart_syscall (); -/* we don't want to modify a running netdev */ -if (netif_running (dev->net)) { -netdev_err (dev->net,"Cannot change a running device\n" ); - ret = -EBUSY ; - goto err; - } - del_dev =qmimux_find_dev (dev, mux_id);if (!del_dev) {netdev_err (dev->net,"mux_id not present\n" );
[email protected]
maintainerPATCH 32/168 netdevsim: init u64 stats for 32bit hardware
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Hillf Danton <[email protected]>
commit 863a42b289c22df63db62b10fc2c2ffc237e2125 upstream.
Init the u64 stats in order to avoid the lockdep prints on the 32bit hardware like
INFO: trying to register non-static key. the code is fine but needs lockdep annotation. turning off the locking correctness validator.
CPU: 0 PID: 4695 Comm: syz-executor.0 Not tainted 5.11.0-rc5-syzkaller #0 Hardware name: ARM-Versatile Express Backtrace: [<826fc5b8>] (dump_backtrace) from [<826fc82c>] (show_stack+0x18/0x1c arch/arm/kernel/traps.c:252) [<826fc814>] (show_stack) from [<8270d1f8>] (__dump_stack lib/dump_stack.c:79 [inline]) [<826fc814>] (show_stack) from [<8270d1f8>] (dump_stack+0xa8/0xc8 lib/dump_stack.c:120) [<8270d150>] (dump_stack) from [<802bf9c0>] (assign_lock_key kernel/locking/lockdep.c:935 [inline]) [<8270d150>] (dump_stack) from [<802bf9c0>] (register_lock_class+0xabc/0xb68 kernel/locking/lockdep.c:1247) [<802bef04>] (register_lock_class) from [<802baa2c>] (__lock_acquire+0x84/0x32d4 kernel/locking/lockdep.c:4711) [<802ba9a8>] (__lock_acquire) from [<802be840>] (lock_acquire.part.0+0xf0/0x554 kernel/locking/lockdep.c:5442) [<802be750>] (lock_acquire.part.0) from [<802bed10>] (lock_acquire+0x6c/0x74 kernel/locking/lockdep.c:5415) [<802beca4>] (lock_acquire) from [<81560548>] (seqcount_lockdep_reader_access include/linux/seqlock.h:103 [inline]) [<802beca4>] (lock_acquire) from [<81560548>] (__u64_stats_fetch_begin include/linux/u64_stats_sync.h:164 [inline]) [<802beca4>] (lock_acquire) from [<81560548>] (u64_stats_fetch_begin include/linux/u64_stats_sync.h:175 [inline]) [<802beca4>] (lock_acquire) from [<81560548>] (nsim_get_stats64+0xdc/0xf0 drivers/net/netdevsim/netdev.c:70) [<8156046c>] (nsim_get_stats64) from [<81e2efa0>] (dev_get_stats+0x44/0xd0 net/core/dev.c:10405) [<81e2ef5c>] (dev_get_stats) from [<81e53204>] (rtnl_fill_stats+0x38/0x120 net/core/rtnetlink.c:1211) [<81e531cc>] (rtnl_fill_stats) from [<81e59d58>] (rtnl_fill_ifinfo+0x6d4/0x148c net/core/rtnetlink.c:1783) [<81e59684>] (rtnl_fill_ifinfo) from [<81e5ceb4>] (rtmsg_ifinfo_build_skb+0x9c/0x108 net/core/rtnetlink.c:3798) [<81e5ce18>] (rtmsg_ifinfo_build_skb) from [<81e5d0ac>] (rtmsg_ifinfo_event net/core/rtnetlink.c:3830 [inline]) [<81e5ce18>] (rtmsg_ifinfo_build_skb) from [<81e5d0ac>] (rtmsg_ifinfo_event net/core/rtnetlink.c:3821 [inline]) [<81e5ce18>] (rtmsg_ifinfo_build_skb) from [<81e5d0ac>] (rtmsg_ifinfo+0x44/0x70 net/core/rtnetlink.c:3839) [<81e5d068>] (rtmsg_ifinfo) from [<81e45c2c>] (register_netdevice+0x664/0x68c net/core/dev.c:10103) [<81e455c8>] (register_netdevice) from [<815608bc>] (nsim_create+0xf8/0x124 drivers/net/netdevsim/netdev.c:317) [<815607c4>] (nsim_create) from [<81561184>] (__nsim_dev_port_add+0x108/0x188 drivers/net/netdevsim/dev.c:941) [<8156107c>] (__nsim_dev_port_add) from [<815620d8>] (nsim_dev_port_add_all drivers/net/netdevsim/dev.c:990 [inline]) [<8156107c>] (__nsim_dev_port_add) from [<815620d8>] (nsim_dev_probe+0x5cc/0x750 drivers/net/netdevsim/dev.c:1119) [<81561b0c>] (nsim_dev_probe) from [<815661dc>] (nsim_bus_probe+0x10/0x14 drivers/net/netdevsim/bus.c:287) [<815661cc>] (nsim_bus_probe) from [<811724c0>] (really_probe+0x100/0x50c drivers/base/dd.c:554) [<811723c0>] (really_probe) from [<811729c4>] (driver_probe_device+0xf8/0x1c8 drivers/base/dd.c:740) [<811728cc>] (driver_probe_device) from [<81172fe4>] (__device_attach_driver+0x8c/0xf0 drivers/base/dd.c:846) [<81172f58>] (__device_attach_driver) from [<8116fee0>] (bus_for_each_drv+0x88/0xd8 drivers/base/bus.c:431) [<8116fe58>] (bus_for_each_drv) from [<81172c6c>] (__device_attach+0xdc/0x1d0 drivers/base/dd.c:914) [<81172b90>] (__device_attach) from [<8117305c>] (device_initial_probe+0x14/0x18 drivers/base/dd.c:961) [<81173048>] (device_initial_probe) from [<81171358>] (bus_probe_device+0x90/0x98 drivers/base/bus.c:491) [<811712c8>] (bus_probe_device) from [<8116e77c>] (device_add+0x320/0x824 drivers/base/core.c:3109) [<8116e45c>] (device_add) from [<8116ec9c>] (device_register+0x1c/0x20 drivers/base/core.c:3182) [<8116ec80>] (device_register) from [<81566710>] (nsim_bus_dev_new drivers/net/netdevsim/bus.c:336 [inline]) [<8116ec80>] (device_register) from [<81566710>] (new_device_store+0x178/0x208 drivers/net/netdevsim/bus.c:215) [<81566598>] (new_device_store) from [<8116fcb4>] (bus_attr_store+0x2c/0x38 drivers/base/bus.c:122) [<8116fc88>] (bus_attr_store) from [<805b4b8c>] (sysfs_kf_write+0x48/0x54 fs/sysfs/file.c:139) [<805b4b44>] (sysfs_kf_write) from [<805b3c90>] (kernfs_fop_write_iter+0x128/0x1ec fs/kernfs/file.c:296) [<805b3b68>] (kernfs_fop_write_iter) from [<804d22fc>] (call_write_iter include/linux/fs.h:1901 [inline]) [<805b3b68>] (kernfs_fop_write_iter) from [<804d22fc>] (new_sync_write fs/read_write.c:518 [inline]) [<805b3b68>] (kernfs_fop_write_iter) from [<804d22fc>] (vfs_write+0x3dc/0x57c fs/read_write.c:605) [<804d1f20>] (vfs_write) from [<804d2604>] (ksys_write+0x68/0xec fs/read_write.c:658) [<804d259c>] (ksys_write) from [<804d2698>] (__do_sys_write fs/read_write.c:670 [inline]) [<804d259c>] (ksys_write) from [<804d2698>] (sys_write+0x10/0x14 fs/read_write.c:667) [<804d2688>] (sys_write) from [<80200060>] (ret_fast_syscall+0x0/0x2c arch/arm/mm/proc-v7.S:64)
--- drivers/net/netdevsim/netdev.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/netdevsim/netdev.c +1 -0
@@ -292,6 +292,7 @@ nsim_create(struct nsim_dev *nsim_dev, s ns =netdev_priv (dev); ns->netdev = dev; +u64_stats_init (&ns->syncp); ns->nsim_dev = nsim_dev; ns->nsim_dev_port = nsim_dev_port; ns->nsim_bus_dev = nsim_dev->nsim_bus_dev;
[email protected]
maintainerPATCH 33/168 cipso,calipso: resolve a number of problems with the DOI refcounts
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Paul Moore <[email protected]>
commit ad5d07f4a9cd671233ae20983848874731102c08 upstream.
The current CIPSO and CALIPSO refcounting scheme for the DOI definitions is a bit flawed in that we:
1. Don't correctly match gets/puts in netlbl_cipsov4_list().
2. Decrement the refcount on each attempt to remove the DOI from the DOI list, only removing it from the list once the refcount drops to zero.
This patch fixes these problems by adding the missing "puts" to netlbl_cipsov4_list() and introduces a more conventional, i.e. not-buggy, refcounting mechanism to the DOI definitions. Upon the addition of a DOI to the DOI list, it is initialized with a refcount of one, removing a DOI from the list removes it from the list and drops the refcount by one; "gets" and "puts" behave as expected with respect to refcounts, increasing and decreasing the DOI's refcount by one.
--- net/ipv4/cipso_ipv4.c | 11 +---------- net/ipv6/calipso.c | 14 +++++--------- net/netlabel/netlabel_cipso_v4.c | 3 +++ 3 files changed, 9 insertions(+), 19 deletions(-)
net/ipv4/cipso_ipv4.c +1 -10
@@ -519,16 +519,10 @@ int cipso_v4_doi_remove(u32 doi, struct ret_val = -ENOENT ; goto doi_remove_return; } -if (!refcount_dec_and_test (&doi_def->refcount)) { -spin_unlock (&cipso_v4_doi_list_lock); - ret_val = -EBUSY ; - goto doi_remove_return; - }list_del_rcu (&doi_def->list);spin_unlock (&cipso_v4_doi_list_lock); -cipso_v4_cache_invalidate (); -call_rcu (&doi_def->rcu, cipso_v4_doi_free_rcu); +cipso_v4_doi_putdef (doi_def); ret_val =0 ; doi_remove_return:
@@ -585,9 +579,6 @@ void cipso_v4_doi_putdef(struct cipso_v4if (!refcount_dec_and_test (&doi_def->refcount))return ; -spin_lock (&cipso_v4_doi_list_lock); -list_del_rcu (&doi_def->list); -spin_unlock (&cipso_v4_doi_list_lock);cipso_v4_cache_invalidate ();call_rcu (&doi_def->rcu, cipso_v4_doi_free_rcu);
net/ipv6/calipso.c +5 -9
@@ -83,6 +83,9 @@ struct calipso_map_cache_entry {static struct calipso_map_cache_bkt *calipso_cache; +static void calipso_cache_invalidate (void ); +static void calipso_doi_putdef (struct calipso_doi *doi_def); +/* Label Mapping Cache Functions */
@@ -444,15 +447,10 @@ static int calipso_doi_remove(u32 doi, s ret_val = -ENOENT ; goto doi_remove_return; } -if (!refcount_dec_and_test (&doi_def->refcount)) { -spin_unlock (&calipso_doi_list_lock); - ret_val = -EBUSY ; - goto doi_remove_return; - }list_del_rcu (&doi_def->list);spin_unlock (&calipso_doi_list_lock); -call_rcu (&doi_def->rcu, calipso_doi_free_rcu); +calipso_doi_putdef (doi_def); ret_val =0 ; doi_remove_return:
@@ -508,10 +506,8 @@ static void calipso_doi_putdef(struct caif (!refcount_dec_and_test (&doi_def->refcount))return ; -spin_lock (&calipso_doi_list_lock); -list_del_rcu (&doi_def->list); -spin_unlock (&calipso_doi_list_lock); +calipso_cache_invalidate ();call_rcu (&doi_def->rcu, calipso_doi_free_rcu); }
net/netlabel/netlabel_cipso_v4.c +3 -0
@@ -575,6 +575,7 @@ list_start:break ; } +cipso_v4_doi_putdef (doi_def);rcu_read_unlock ();genlmsg_end (ans_skb, data);
@@ -583,12 +584,14 @@ list_start: list_retry:/* XXX - this limit is a guesstimate */ if (nlsze_mult <4 ) { +cipso_v4_doi_putdef (doi_def);rcu_read_unlock ();kfree_skb (ans_skb); nlsze_mult *=2 ; goto list_start; } list_failure_lock: +cipso_v4_doi_putdef (doi_def);rcu_read_unlock (); list_failure:kfree_skb (ans_skb);
[email protected]
maintainerPATCH 34/168 net: lapbether: Remove netif_start_queue / netif_stop_queue
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Xie He <[email protected]>
commit f7d9d4854519fdf4d45c70a4d953438cd88e7e58 upstream.
For the devices in this driver, the default qdisc is "noqueue", because their "tx_queue_len" is 0.
In function "__dev_queue_xmit" in "net/core/dev.c", devices with the "noqueue" qdisc are specially handled. Packets are transmitted without being queued after a "dev->flags & IFF_UP" check. However, it's possible that even if this check succeeds, "ops->ndo_stop" may still have already been called. This is because in "__dev_close_many", "ops->ndo_stop" is called before clearing the "IFF_UP" flag.
If we call "netif_stop_queue" in "ops->ndo_stop", then it's possible in "__dev_queue_xmit", it sees the "IFF_UP" flag is present, and then it checks "netif_xmit_stopped" and finds that the queue is already stopped. In this case, it will complain that:
"Virtual device ... asks to queue packet!"
To prevent "__dev_queue_xmit" from generating this complaint, we should not call "netif_stop_queue" in "ops->ndo_stop".
We also don't need to call "netif_start_queue" in "ops->ndo_open", because after a netdev is allocated and registered, the "__QUEUE_STATE_DRV_XOFF" flag is initially not set, so there is no need to call "netif_start_queue" to clear it.
--- drivers/net/wan/lapbether.c | 3 --- 1 file changed, 3 deletions(-)
drivers/net/wan/lapbether.c +0 -3
@@ -283,7 +283,6 @@ static int lapbeth_open(struct net_devicreturn -ENODEV ; } -netif_start_queue (dev);return 0 ; }
@@ -291,8 +290,6 @@ static int lapbeth_close(struct net_devi {int err; -netif_stop_queue (dev); -if ((err =lapb_unregister (dev)) !=LAPB_OK )pr_err ("lapb_unregister error: %d\n" , err);
[email protected]
maintainerPATCH 35/168 net: davicom: Fix regulator not turned off on failed probe
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Paul Cercueil <[email protected]>
commit ac88c531a5b38877eba2365a3f28f0c8b513dc33 upstream.
When the probe fails or requests to be defered, we must disable the regulator that was previously enabled.
--- drivers/net/ethernet/davicom/dm9000.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
drivers/net/ethernet/davicom/dm9000.c +9 -3
@@ -1454,7 +1454,7 @@ dm9000_probe(struct platform_device *pdeif (ret) {dev_err (dev,"failed to request reset gpio %d: %d\n" , reset_gpios, ret); -return -ENODEV ; + goto out_regulator_disable; }/* According to manual PWRST# Low Period Min 1ms */
@@ -1466,8 +1466,10 @@ dm9000_probe(struct platform_device *pdeif (!pdata) { pdata =dm9000_parse_dt (&pdev->dev); -if (IS_ERR (pdata)) -return PTR_ERR (pdata); +if (IS_ERR (pdata)) { + ret =PTR_ERR (pdata); + goto out_regulator_disable; + } }/* Init network device */
@@ -1708,6 +1710,10 @@ out:dm9000_release_board (pdev, db);free_netdev (ndev); +out_regulator_disable: +if (!IS_ERR (power)) +regulator_disable (power); +return ret; }
[email protected]
maintainerPATCH 36/168 net: davicom: Fix regulator not turned off on driver removal
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Paul Cercueil <[email protected]>
commit cf9e60aa69ae6c40d3e3e4c94dd6c8de31674e9b upstream.
We must disable the regulator that was enabled in the probe function.
--- drivers/net/ethernet/davicom/dm9000.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/net/ethernet/davicom/dm9000.c +8 -1
@@ -134,6 +134,8 @@ struct board_info {u32 wake_state;int ip_summed; + +struct regulator *power_supply; };/* debug code */
@@ -1486,6 +1488,8 @@ dm9000_probe(struct platform_device *pde db->dev = &pdev->dev; db->ndev = ndev; +if (!IS_ERR (power)) + db->power_supply = power;spin_lock_init (&db->lock);mutex_init (&db->addr_lock);
@@ -1771,10 +1775,13 @@ static int dm9000_drv_remove(struct platform_device *pdev) {struct net_device *ndev =platform_get_drvdata (pdev); +struct board_info *dm =to_dm9000_board (ndev);unregister_netdev (ndev); -dm9000_release_board (pdev,netdev_priv (ndev)); +dm9000_release_board (pdev, dm);free_netdev (ndev);/* free device structure */ +if (dm->power_supply) +regulator_disable (dm->power_supply);dev_dbg (&pdev->dev,"released and freed device\n" );return 0 ;
[email protected]
maintainerPATCH 37/168 net: qrtr: fix error return code of qrtr_sendmsg()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Jia-Ju Bai <[email protected]>
commit 179d0ba0c454057a65929c46af0d6ad986754781 upstream.
When sock_alloc_send_skb() returns NULL to skb, no error return code of qrtr_sendmsg() is assigned.
To fix this bug, rc is assigned with -ENOMEM in this case.
--- net/qrtr/qrtr.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
net/qrtr/qrtr.c +3 -1
@@ -791,8 +791,10 @@ static int qrtr_sendmsg(struct socket *s plen = (len +3 ) & ~3 ; skb =sock_alloc_send_skb (sk, plen +QRTR_HDR_MAX_SIZE , msg->msg_flags &MSG_DONTWAIT , &rc); -if (!skb) +if (!skb) { + rc = -ENOMEM ; goto out_node; + }skb_reserve (skb,QRTR_HDR_MAX_SIZE );
[email protected]
maintainerPATCH 38/168 ixgbe: fail to create xfrm offload of IPsec tunnel mode SA
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Antony Antony <[email protected]>
commit d785e1fec60179f534fbe8d006c890e5ad186e51 upstream.
Based on talks and indirect references ixgbe IPsec offlod do not support IPsec tunnel mode offload. It can only support IPsec transport mode offload. Now explicitly fail when creating non transport mode SA with offload to avoid false performance expectations.
--- drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c | 5 +++++ drivers/net/ethernet/intel/ixgbevf/ipsec.c | 5 +++++ 2 files changed, 10 insertions(+)
drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c +5 -0
@@ -575,6 +575,11 @@ static int ixgbe_ipsec_add_sa(struct xfrreturn -EINVAL ; } +if (xs->props.mode !=XFRM_MODE_TRANSPORT ) { +netdev_err (dev,"Unsupported mode for ipsec offload\n" ); +return -EINVAL ; + } +if (ixgbe_ipsec_check_mgmt_ip (xs)) {netdev_err (dev,"IPsec IP addr clash with mgmt filters\n" );return -EINVAL ;
drivers/net/ethernet/intel/ixgbevf/ipsec.c +5 -0
@@ -272,6 +272,11 @@ static int ixgbevf_ipsec_add_sa(struct xreturn -EINVAL ; } +if (xs->props.mode !=XFRM_MODE_TRANSPORT ) { +netdev_err (dev,"Unsupported mode for ipsec offload\n" ); +return -EINVAL ; + } +if (xs->xso.flags &XFRM_OFFLOAD_INBOUND ) {struct rx_sa rsa;
[email protected]
maintainerPATCH 39/168 net: stmmac: stop each tx channel independently
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Joakim Zhang <[email protected]>
commit a3e860a83397bf761ec1128a3f0ba186445992c6 upstream.
If clear GMAC_CONFIG_TE bit, it would stop all tx channels, but users may only want to stop specific tx channel.
--- drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c | 4 ---- 1 file changed, 4 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c +0 -4
@@ -60,10 +60,6 @@ void dwmac4_dma_stop_tx(void __iomem *io value &= ~DMA_CONTROL_ST ;writel (value, ioaddr +DMA_CHAN_TX_CONTROL (chan)); - - value =readl (ioaddr +GMAC_CONFIG ); - value &= ~GMAC_CONFIG_TE ; -writel (value, ioaddr +GMAC_CONFIG ); }void dwmac4_dma_start_rx (void __iomem *ioaddr,u32 chan)
[email protected]
maintainerPATCH 40/168 net: stmmac: fix watchdog timeout during suspend/resume stress test
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Joakim Zhang <[email protected]>
commit c511819d138de38e1637eedb645c207e09680d0f upstream.
stmmac_xmit() call stmmac_tx_timer_arm() at the end to modify tx timer to do the transmission cleanup work. Imagine such a situation, stmmac enters suspend immediately after tx timer modified, it's expire callback stmmac_tx_clean() would not be invoked. This could affect BQL, since netdev_tx_sent_queue() has been called, but netdev_tx_completed_queue() have not been involved, as a result, dql_avail(&dev_queue->dql) finally always return a negative value.
__dev_queue_xmit->__dev_xmit_skb->qdisc_run->__qdisc_run->qdisc_restart->dequeue_skb:
if ((q->flags & TCQ_F_ONETXQUEUE) && netif_xmit_frozen_or_stopped(txq)) // __QUEUE_STATE_STACK_XOFF is set
Net core will stop transmitting any more. Finillay, net watchdong would timeout. To fix this issue, we should call netdev_tx_reset_queue() in stmmac_resume().
--- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +2 -0
@@ -4821,6 +4821,8 @@ static void stmmac_reset_queues_param(st tx_q->cur_tx =0 ; tx_q->dirty_tx =0 ; tx_q->mss =0 ; + +netdev_tx_reset_queue (netdev_get_tx_queue (priv->dev, queue)); } }
[email protected]
maintainerPATCH 41/168 selftests: forwarding: Fix race condition in mirror installation
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Danielle Ratson <[email protected]>
commit edcbf5137f093b5502f5f6b97cce3cbadbde27aa upstream.
When mirroring to a gretap in hardware the device expects to be programmed with the egress port and all the encapsulating headers. This requires the driver to resolve the path the packet will take in the software data path and program the device accordingly.
If the path cannot be resolved (in this case because of an unresolved neighbor), then mirror installation fails until the path is resolved. This results in a race that causes the test to sometimes fail.
Fix this by setting the neighbor's state to permanent, so that it is always valid.
--- tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d_vlan.sh | 9 +++++++++ 1 file changed, 9 insertions(+)
tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d_vlan.sh +9 -0
@@ -86,11 +86,20 @@ test_ip6gretap() test_gretap_stp() { + # Sometimes after mirror installation, the neighbor's state is not valid. + # The reason is that there is no SW datapath activity related to the + # neighbor for the remote GRE address. Therefore whether the corresponding + # neighbor will be valid is a matter of luck, and the test is thus racy. + # Set the neighbor's state to permanent, so it would be always valid. + ip neigh replace 192.0.2.130 lladdr $(mac_get $h3) \ + nud permanent dev br2 full_test_span_gre_stp gt4 $swp3.555 "mirror to gretap" } test_ip6gretap_stp() { + ip neigh replace 2001:db8:2::2 lladdr $(mac_get $h3) \ + nud permanent dev br2 full_test_span_gre_stp gt6 $swp3.555 "mirror to ip6gretap" }
[email protected]
maintainerPATCH 42/168 perf traceevent: Ensure read cmdlines are null terminated.
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Rogers <[email protected]>
commit 137a5258939aca56558f3a23eb229b9c4b293917 upstream.
Issue detected by address sanitizer.
--- tools/perf/util/trace-event-read.c | 1 + 1 file changed, 1 insertion(+)
tools/perf/util/trace-event-read.c +1 -0
@@ -361,6 +361,7 @@ static int read_saved_cmdline(struct teppr_debug ("error reading saved cmdlines\n" ); goto out; } + buf[ret] ='\0' ;parse_saved_cmdline (pevent, buf, size); ret =0 ;
[email protected]
maintainerPATCH 43/168 net: hns3: fix query vlan mask value error for flow director
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Jian Shen <[email protected]>
commit c75ec148a316e8cf52274d16b9b422703b96f5ce upstream.
Currently, the driver returns VLAN_VID_MASK for vlan mask field, when get flow director rule information for rule doesn't use vlan. It may cause the vlan mask value display as 0xf000 in this case, like below:
estuary:/$ ethtool -u eth1 50 RX rings available Total 1 rules
Filter: 2
Rule Type: TCP over IPv4
Src IP addr: 0.0.0.0 mask: 255.255.255.255
Dest IP addr: 0.0.0.0 mask: 255.255.255.255
TOS: 0x0 mask: 0xff
Src port: 0 mask: 0xffff
Dest port: 0 mask: 0xffff
VLAN EtherType: 0x0 mask: 0xffff
VLAN: 0x0 mask: 0xf000
User-defined: 0x1234 mask: 0x0
Action: Direct to queue 3
Fix it by return 0.
--- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +1 -2
@@ -5939,8 +5939,7 @@ static int hclge_get_fd_rule_info(struct fs->h_ext.vlan_tci =cpu_to_be16 (rule->tuples.vlan_tag1); fs->m_ext.vlan_tci = rule->unused_tuple &BIT (INNER_VLAN_TAG_FST ) ? -cpu_to_be16 (VLAN_VID_MASK ) : -cpu_to_be16 (rule->tuples_mask.vlan_tag1); +0 :cpu_to_be16 (rule->tuples_mask.vlan_tag1); }if (fs->flow_type &FLOW_MAC_EXT ) {
[email protected]
maintainerPATCH 44/168 net: hns3: fix bug when calculating the TCAM table info
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Jian Shen <[email protected]>
commit b36fc875bcdee56865c444a2cdae17d354a6d5f5 upstream.
The function hclge_fd_convert_tuple() is used to convert tuples and tuples mask to TCAM x and y. But it misuses the source mac as source mac mask when convert INNER_SRC_MAC, which may cause the flow director rule works unexpectedly. So fix it.
--- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +2 -2
@@ -4908,9 +4908,9 @@ static bool hclge_fd_convert_tuple(u32 tcase BIT (INNER_SRC_MAC ):for (i =0 ; i <ETH_ALEN ; i++) {calc_x (key_x[ETH_ALEN -1 - i], rule->tuples.src_mac[i], - rule->tuples.src_mac[i]); + rule->tuples_mask.src_mac[i]);calc_y (key_y[ETH_ALEN -1 - i], rule->tuples.src_mac[i], - rule->tuples.src_mac[i]); + rule->tuples_mask.src_mac[i]); }return true;
[email protected]
maintainerPATCH 45/168 s390/cio: return -EFAULT if copy_to_user() fails again
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Wang Qing <[email protected]>
commit 51c44babdc19aaf882e1213325a0ba291573308f upstream.
The copy_to_user() function returns the number of bytes remaining to be copied, but we want to return -EFAULT if the copy doesn't complete.
--- drivers/s390/cio/vfio_ccw_ops.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/s390/cio/vfio_ccw_ops.c +2 -2
@@ -506,7 +506,7 @@ static ssize_t vfio_ccw_mdev_ioctl(strucif (ret)return ret; -return copy_to_user ((void __user *)arg, &info, minsz); +return copy_to_user ((void __user *)arg, &info, minsz) ? -EFAULT :0 ; }case VFIO_DEVICE_GET_REGION_INFO : {
@@ -524,7 +524,7 @@ static ssize_t vfio_ccw_mdev_ioctl(strucif (ret)return ret; -return copy_to_user ((void __user *)arg, &info, minsz); +return copy_to_user ((void __user *)arg, &info, minsz) ? -EFAULT :0 ; }case VFIO_DEVICE_GET_IRQ_INFO : {
[email protected]
maintainerPATCH 46/168 bnxt_en: reliably allocate IRQ table on reset to avoid crash
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Edwin Peer <[email protected]>
commit 20d7d1c5c9b11e9f538ed4a2289be106de970d3e upstream.
The following trace excerpt corresponds with a NULL pointer dereference of 'bp->irq_tbl' in bnxt_setup_inta() on an Aarch64 system after many device resets:
Unable to handle kernel NULL pointer dereference at ... 000000d
...
pc : string+0x3c/0x80
lr : vsnprintf+0x294/0x7e0
sp : ffff00000f61ba70 pstate : 20000145
x29: ffff00000f61ba70 x28: 000000000000000d
x27: ffff0000009c8b5a x26: ffff00000f61bb80
x25: ffff0000009c8b5a x24: 0000000000000012
x23: 00000000ffffffe0 x22: ffff000008990428
x21: ffff00000f61bb80 x20: 000000000000000d
x19: 000000000000001f x18: 0000000000000000
x17: 0000000000000000 x16: ffff800b6d0fb400
x15: 0000000000000000 x14: ffff800b7fe31ae8
x13: 00001ed16472c920 x12: ffff000008c6b1c9
x11: ffff000008cf0580 x10: ffff00000f61bb80
x9 : 00000000ffffffd8 x8 : 000000000000000c
x7 : ffff800b684b8000 x6 : 0000000000000000
x5 : 0000000000000065 x4 : 0000000000000001
x3 : ffff0a00ffffff04 x2 : 000000000000001f
x1 : 0000000000000000 x0 : 000000000000000d
Call trace:
string+0x3c/0x80
vsnprintf+0x294/0x7e0
snprintf+0x44/0x50
__bnxt_open_nic+0x34c/0x928 [bnxt_en]
bnxt_open+0xe8/0x238 [bnxt_en]
__dev_open+0xbc/0x130
__dev_change_flags+0x12c/0x168
dev_change_flags+0x20/0x60
...Ordinarily, a call to bnxt_setup_inta() (not in trace due to inlining) would not be expected on a system supporting MSIX at all. However, if bnxt_init_int_mode() does not end up being called after the call to bnxt_clear_int_mode() in bnxt_fw_reset_close(), then the driver will think that only INTA is supported and bp->irq_tbl will be NULL, causing the above crash.
In the error recovery scenario, we call bnxt_clear_int_mode() in bnxt_fw_reset_close() early in the sequence. Ordinarily, we will call bnxt_init_int_mode() in bnxt_hwrm_if_change() after we reestablish communication with the firmware after reset. However, if the sequence has to abort before we call bnxt_init_int_mode() and if the user later attempts to re-open the device, then it will cause the crash above.
We fix it in 2 ways:
1. Check for bp->irq_tbl in bnxt_setup_int_mode(). If it is NULL, call bnxt_init_init_mode().
2. If we need to abort in bnxt_hwrm_if_change() and cannot complete the error recovery sequence, set the BNXT_STATE_ABORT_ERR flag. This will cause more drastic recovery at the next attempt to re-open the device, including a call to bnxt_init_int_mode().
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +12 -2
@@ -7925,10 +7925,18 @@ static void bnxt_setup_inta(struct bnxt bp->irq_tbl[0 ].handler = bnxt_inta; } +static int bnxt_init_int_mode (struct bnxt *bp); +static int bnxt_setup_int_mode (struct bnxt *bp) {int rc; +if (!bp->irq_tbl) { + rc =bnxt_init_int_mode (bp); +if (rc || !bp->irq_tbl) +return rc ?: -ENODEV ; + } +if (bp->flags &BNXT_FLAG_USING_MSIX )bnxt_setup_msix (bp);else
@@ -8113,7 +8121,7 @@ static int bnxt_init_inta(struct bnxt *bstatic int bnxt_init_int_mode (struct bnxt *bp) { -int rc =0 ; +int rc = -ENODEV ;if (bp->flags &BNXT_FLAG_MSIX_CAP ) rc =bnxt_init_msix (bp);
@@ -8748,7 +8756,8 @@ static int bnxt_hwrm_if_change(struct bn {struct hwrm_func_drv_if_change_output *resp = bp->hwrm_cmd_resp_addr;struct hwrm_func_drv_if_change_input req = {0 }; -bool resc_reinit = false, fw_reset = false; +bool fw_reset = !bp->irq_tbl; +bool resc_reinit = false;u32 flags =0 ;int rc;
@@ -8776,6 +8785,7 @@ static int bnxt_hwrm_if_change(struct bnif (test_bit (BNXT_STATE_IN_FW_RESET , &bp->state) && !fw_reset) {netdev_err (bp->dev,"RESET_DONE not set during FW reset.\n" ); +set_bit (BNXT_STATE_ABORT_ERR , &bp->state);return -ENODEV ; }if (resc_reinit || fw_reset) {
[email protected]
maintainerPATCH 47/168 drm/compat: Clear bounce structures
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Daniel Vetter <[email protected]>
commit de066e116306baf3a6a62691ac63cfc0b1dabddb upstream.
Some of them have gaps, or fields we don't clear. Native ioctl code does full copies plus zero-extends on size mismatch, so nothing can leak. But compat is more hand-rolled so need to be careful.
None of these matter for performance, so just memset.
Also I didn't fix up the CONFIG_DRM_LEGACY or CONFIG_DRM_AGP ioctl, those are security holes anyway.
--- drivers/gpu/drm/drm_ioc32.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
drivers/gpu/drm/drm_ioc32.c +11 -0
@@ -99,6 +99,8 @@ static int compat_drm_version(struct filif (copy_from_user (&v32, (void __user *)arg,sizeof (v32)))return -EFAULT ; +memset (&v,0 ,sizeof (v)); + v = (struct drm_version ) { .name_len = v32.name_len, .name =compat_ptr (v32.name),
@@ -137,6 +139,9 @@ static int compat_drm_getunique(struct fif (copy_from_user (&uq32, (void __user *)arg,sizeof (uq32)))return -EFAULT ; + +memset (&uq,0 ,sizeof (uq)); + uq = (struct drm_unique ){ .unique_len = uq32.unique_len, .unique =compat_ptr (uq32.unique),
@@ -265,6 +270,8 @@ static int compat_drm_getclient(struct fif (copy_from_user (&c32, argp,sizeof (c32)))return -EFAULT ; +memset (&client,0 ,sizeof (client)); + client.idx = c32.idx; err =drm_ioctl_kernel (file, drm_getclient, &client,0 );
@@ -850,6 +857,8 @@ static int compat_drm_wait_vblank(structif (copy_from_user (&req32, argp,sizeof (req32)))return -EFAULT ; +memset (&req,0 ,sizeof (req)); + req.request.type = req32.request.type; req.request.sequence = req32.request.sequence; req.request.signal = req32.request.signal;
@@ -887,6 +896,8 @@ static int compat_drm_mode_addfb2(structstruct drm_mode_fb_cmd2 req64;int err; +memset (&req64,0 ,sizeof (req64)); +if (copy_from_user (&req64, argp, offsetof(drm_mode_fb_cmd232_t , modifier)))return -EFAULT ;
[email protected]
maintainerPATCH 48/168 drm/shmem-helper: Check for purged buffers in fault handler
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Neil Roberts <[email protected]>
commit d611b4a0907cece060699f2fd347c492451cd2aa upstream.
When a buffer is madvised as not needed and then purged, any attempts to access the buffer from user-space should cause a bus fault. This patch adds a check for that.
--- drivers/gpu/drm/drm_gem_shmem_helper.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-)
drivers/gpu/drm/drm_gem_shmem_helper.c +14 -4
@@ -474,14 +474,24 @@ static vm_fault_t drm_gem_shmem_fault(ststruct drm_gem_object *obj = vma->vm_private_data;struct drm_gem_shmem_object *shmem =to_drm_gem_shmem_obj (obj);loff_t num_pages = obj->size >>PAGE_SHIFT ; +vm_fault_t ret;struct page *page; -if (vmf->pgoff >= num_pages ||WARN_ON_ONCE (!shmem->pages)) -return VM_FAULT_SIGBUS ; +mutex_lock (&shmem->pages_lock); - page = shmem->pages[vmf->pgoff]; +if (vmf->pgoff >= num_pages || +WARN_ON_ONCE (!shmem->pages) || + shmem->madv <0 ) { + ret =VM_FAULT_SIGBUS ; + }else { + page = shmem->pages[vmf->pgoff]; -return vmf_insert_page (vma, vmf->address, page); + ret =vmf_insert_page (vma, vmf->address, page); + } + +mutex_unlock (&shmem->pages_lock); + +return ret; }static void drm_gem_shmem_vm_open (struct vm_area_struct *vma)
[email protected]
maintainerPATCH 49/168 drm/shmem-helper: Dont remove the offset in vm_area_struct pgoff
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Neil Roberts <[email protected]>
commit 11d5a4745e00e73745774671dbf2fb07bd6e2363 upstream.
When mmapping the shmem, it would previously adjust the pgoff in the vm_area_struct to remove the fake offset that is added to be able to identify the buffer. This patch removes the adjustment and makes the fault handler use the vm_fault address to calculate the page offset instead. Although using this address is apparently discouraged, several DRM drivers seem to be doing it anyway.
The problem with removing the pgoff is that it prevents drm_vma_node_unmap from working because that searches the mapping tree by address. That doesn't work because all of the mappings are at offset
0. drm_vma_node_unmap is being used by the shmem helpers when purging the buffer.
This fixes a bug in Panfrost which is using drm_gem_shmem_purge. Without this the mapping for the purged buffer can still be accessed which might mean it would access random pages from other buffers
v2: Don't check whether the unsigned page_offset is less than 0.
--- drivers/gpu/drm/drm_gem_shmem_helper.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/gpu/drm/drm_gem_shmem_helper.c +6 -5
@@ -476,15 +476,19 @@ static vm_fault_t drm_gem_shmem_fault(stloff_t num_pages = obj->size >>PAGE_SHIFT ;vm_fault_t ret;struct page *page; +pgoff_t page_offset; + +/* We don't use vmf->pgoff since that has the fake offset */ + page_offset = (vmf->address - vma->vm_start) >>PAGE_SHIFT ;mutex_lock (&shmem->pages_lock); -if (vmf->pgoff >= num_pages || +if (page_offset >= num_pages ||WARN_ON_ONCE (!shmem->pages) || shmem->madv <0 ) { ret =VM_FAULT_SIGBUS ; }else { - page = shmem->pages[vmf->pgoff]; + page = shmem->pages[page_offset]; ret =vmf_insert_page (vma, vmf->address, page); }
@@ -559,9 +563,6 @@ int drm_gem_shmem_mmap(struct file *filp vma->vm_flags &= ~VM_PFNMAP ; vma->vm_flags |=VM_MIXEDMAP ; -/* Remove the fake offset */ - vma->vm_pgoff -=drm_vma_node_start (&shmem->base.vma_node); -return 0 ; }EXPORT_SYMBOL_GPL (drm_gem_shmem_mmap);
[email protected]
maintainerPATCH 50/168 drm: meson_drv add shutdown function
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Artem Lapkin <[email protected]>
commit fa0c16caf3d73ab4d2e5d6fa2ef2394dbec91791 upstream.
Problem: random stucks on reboot stage about 1/20 stuck/reboots // debug kernel log
[ 4.496660] reboot: kernel restart prepare CMD:(null) [ 4.498114] meson_ee_pwrc c883c000.system-controller:power-controller: shutdown begin [ 4.503949] meson_ee_pwrc c883c000.system-controller:power-controller: shutdown domain 0:VPU...
...STUCK...
Solution: add shutdown function to meson_drm driver // debug kernel log
[ 5.231896] reboot: kernel restart prepare CMD:(null) [ 5.246135] [drm:meson_drv_shutdown] ... [ 5.259271] meson_ee_pwrc c883c000.system-controller:power-controller: shutdown begin [ 5.274688] meson_ee_pwrc c883c000.system-controller:power-controller: shutdown domain 0:VPU... [ 5.338331] reboot: Restarting system [ 5.358293] psci: PSCI_0_2_FN_SYSTEM_RESET reboot_mode:0 cmd:(null)
bl31 reboot reason: 0xd
bl31 reboot reason: 0x0
system cmd 1.
...REBOOT...
Tested: on VIM1 VIM2 VIM3 VIM3L khadas sbcs - 1000+ successful reboots and Odroid boards, WeTek Play2 (GXBB)
--- drivers/gpu/drm/meson/meson_drv.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
drivers/gpu/drm/meson/meson_drv.c +11 -0
@@ -420,6 +420,16 @@ static int meson_probe_remote(struct plareturn count; } +static void meson_drv_shutdown (struct platform_device *pdev) +{ +struct meson_drm *priv =dev_get_drvdata (&pdev->dev); +struct drm_device *drm = priv->drm; + +DRM_DEBUG_DRIVER ("\n" ); +drm_kms_helper_poll_fini (drm); +drm_atomic_helper_shutdown (drm); +} +static int meson_drv_probe (struct platform_device *pdev) {struct component_match *match =NULL ;
@@ -469,6 +479,7 @@ MODULE_DEVICE_TABLE(of, dt_match);static struct platform_driver meson_drm_platform_driver = { .probe = meson_drv_probe, + .shutdown = meson_drv_shutdown, .driver = { .name ="meson-drm" , .of_match_table = dt_match,
[email protected]
maintainerPATCH 51/168 s390/cio: return -EFAULT if copy_to_user() fails
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Eric Farman <[email protected]>
commit d9c48a948d29bcb22f4fe61a81b718ef6de561a0 upstream.
--- drivers/s390/cio/vfio_ccw_ops.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/s390/cio/vfio_ccw_ops.c +1 -1
@@ -545,7 +545,7 @@ static ssize_t vfio_ccw_mdev_ioctl(strucif (info.count ==-1 )return -EINVAL ; -return copy_to_user ((void __user *)arg, &info, minsz); +return copy_to_user ((void __user *)arg, &info, minsz) ? -EFAULT :0 ; }case VFIO_DEVICE_SET_IRQS : {
[email protected]
maintainerPATCH 52/168 s390/crypto: return -EFAULT if copy_to_user() fails
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Wang Qing <[email protected]>
commit 942df4be7ab40195e2a839e9de81951a5862bc5b upstream.
The copy_to_user() function returns the number of bytes remaining to be copied, but we want to return -EFAULT if the copy doesn't complete.
--- drivers/s390/crypto/vfio_ap_ops.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/s390/crypto/vfio_ap_ops.c +1 -1
@@ -1279,7 +1279,7 @@ static int vfio_ap_mdev_get_device_info( info.num_regions =0 ; info.num_irqs =0 ; -return copy_to_user ((void __user *)arg, &info, minsz); +return copy_to_user ((void __user *)arg, &info, minsz) ? -EFAULT :0 ; } static ssize_t vfio_ap_mdev_ioctl(struct mdev_device *mdev,
[email protected]
maintainerPATCH 53/168 qxl: Fix uninitialised struct field head.surface_id
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Colin Ian King <[email protected]>
commit 738acd49eb018feb873e0fac8f9517493f6ce2c7 upstream.
The surface_id struct field in head is not being initialized and static analysis warns that this is being passed through to dev->monitors_config->heads[i] on an assignment. Clear up this warning by initializing it to zero.
Addresses-Coverity: ("Uninitialized scalar variable")
--- drivers/gpu/drm/qxl/qxl_display.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/qxl/qxl_display.c +1 -0
@@ -325,6 +325,7 @@ static void qxl_crtc_update_monitors_con head.id = i; head.flags =0 ; + head.surface_id =0 ; oldcount = qdev->monitors_config->count;if (crtc->state->active) {struct drm_display_mode *mode = &crtc->mode;
[email protected]
maintainerPATCH 54/168 sh_eth: fix TRSCER mask for R7S9210
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Sergey Shtylyov <[email protected]>
commit 165bc5a4f30eee4735845aa7dbd6b738643f2603 upstream.
According to the RZ/A2M Group User's Manual: Hardware, Rev. 2.00, the TRSCER register has bit 9 reserved, hence we can't use the driver's default TRSCER mask. Add the explicit initializer for sh_eth_cpu_data::trscer_err_mask for R7S9210.
--- drivers/net/ethernet/renesas/sh_eth.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/renesas/sh_eth.c +2 -0
@@ -828,6 +828,8 @@ static struct sh_eth_cpu_data r7s9210_da .fdr_value = 0x0000070f, + .trscer_err_mask = DESC_I_RINT8 | DESC_I_RINT5, + .apr = 1, .mpr = 1, .tpauser = 1,
[email protected]
maintainerPATCH 55/168 media: usbtv: Fix deadlock on suspend
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Maxim Mikityanskiy <[email protected]>
commit 8a7e27fd5cd696ba564a3f62cedef7269cfd0723 upstream.
usbtv doesn't support power management, so on system suspend the .disconnect callback of the driver is called. The teardown sequence includes a call to snd_card_free. Its implementation waits until the refcount of the sound card device drops to zero, however, if its file is open, snd_card_file_add takes a reference, which can't be dropped during the suspend, because the userspace processes are already frozen at this point. snd_card_free waits for completion forever, leading to a hang on suspend.
This commit fixes this deadlock condition by replacing snd_card_free with snd_card_free_when_closed, that doesn't wait until all references are released, allowing suspend to progress.
--- drivers/media/usb/usbtv/usbtv-audio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/media/usb/usbtv/usbtv-audio.c +1 -1
@@ -399,7 +399,7 @@ void usbtv_audio_free(struct usbtv *usbtcancel_work_sync (&usbtv->snd_trigger);if (usbtv->snd && usbtv->udev) { -snd_card_free (usbtv->snd); +snd_card_free_when_closed (usbtv->snd); usbtv->snd =NULL ; } }
[email protected]
maintainerPATCH 56/168 media: v4l: vsp1: Fix uif null pointer access
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Biju Das <[email protected]>
commit 6732f313938027a910e1f7351951ff52c0329e70 upstream.
RZ/G2L SoC has no UIF. This patch fixes null pointer access, when UIF module is not used.
--- drivers/media/platform/vsp1/vsp1_drm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/media/platform/vsp1/vsp1_drm.c +2 -2
@@ -462,9 +462,9 @@ static int vsp1_du_pipeline_setup_inputs* make sure it is present in the pipeline's list of entities if it * wasn't already. */ -if (!use_uif) { +if (drm_pipe->uif && !use_uif) { drm_pipe->uif->pipe =NULL ; - }else if (!drm_pipe->uif->pipe) { + }else if (drm_pipe->uif && !drm_pipe->uif->pipe) { drm_pipe->uif->pipe = pipe;list_add_tail (&drm_pipe->uif->list_pipe, &pipe->entities); }
[email protected]
maintainerPATCH 57/168 media: v4l: vsp1: Fix bru null pointer access
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Biju Das <[email protected]>
commit ac8d82f586c8692b501cb974604a71ef0e22a04c upstream.
RZ/G2L SoC has only BRS. This patch fixes null pointer access,when only BRS is enabled.
--- drivers/media/platform/vsp1/vsp1_drm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/media/platform/vsp1/vsp1_drm.c +1 -1
@@ -245,7 +245,7 @@ static int vsp1_du_pipeline_setup_brx(st brx = &vsp1->bru->entity; else if (pipe->brx && !drm_pipe->force_brx_release) brx = pipe->brx; - else if (!vsp1->bru->entity.pipe) + else if (vsp1_feature(vsp1, VSP1_HAS_BRU) && !vsp1->bru->entity.pipe) brx = &vsp1->bru->entity; else brx = &vsp1->brs->entity;
[email protected]
maintainerPATCH 58/168 media: rc: compile rc-cec.c into rc-core
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Hans Verkuil <[email protected]>
commit f09f9f93afad770a04b35235a0aa465fcc8d6e3d upstream.
The rc-cec keymap is unusual in that it can't be built as a module, instead it is registered directly in rc-main.c if CONFIG_MEDIA_CEC_RC is set. This is because it can be called from drm_dp_cec_set_edid() via cec_register_adapter() in an asynchronous context, and it is not allowed to use request_module() to load rc-cec.ko in that case. Trying to do so results in a 'WARN_ON_ONCE(wait && current_is_async())'.
Since this keymap is only used if CONFIG_MEDIA_CEC_RC is set, we just compile this keymap into the rc-core module and never as a separate module.
--- drivers/media/rc/Makefile | 1 + drivers/media/rc/keymaps/Makefile | 1 - drivers/media/rc/keymaps/rc-cec.c | 28 +++++++++++----------------- drivers/media/rc/rc-main.c | 6 ++++++ include/media/rc-map.h | 7 +++++++ 5 files changed, 25 insertions(+), 18 deletions(-)
drivers/media/rc/Makefile +1 -0
@@ -5,6 +5,7 @@ obj-y += keymaps/ obj-$(CONFIG_RC_CORE) += rc-core.o rc-core-y := rc-main.o rc-ir-raw.o rc-core-$(CONFIG_LIRC) += lirc_dev.o +rc-core-$(CONFIG_MEDIA_CEC_RC) += keymaps/rc-cec.o rc-core-$(CONFIG_BPF_LIRC_MODE2) += bpf-lirc.o obj-$(CONFIG_IR_NEC_DECODER) += ir-nec-decoder.o obj-$(CONFIG_IR_RC5_DECODER) += ir-rc5-decoder.o
drivers/media/rc/keymaps/Makefile +0 -1
@@ -20,7 +20,6 @@ obj-$(CONFIG_RC_MAP) += rc-adstech-dvb-t rc-behold.o \ rc-behold-columbus.o \ rc-budget-ci-old.o \ - rc-cec.o \ rc-cinergy-1400.o \ rc-cinergy.o \ rc-d680-dmb.o \
drivers/media/rc/keymaps/rc-cec.c +11 -17
@@ -1,6 +1,16 @@// SPDX-License-Identifier: GPL-2.0-or-later /* Keytable for the CEC remote control * +* This keymap is unusual in that it can't be built as a module, +* instead it is registered directly in rc-main.c if CONFIG_MEDIA_CEC_RC +* is set. This is because it can be called from drm_dp_cec_set_edid() via +* cec_register_adapter() in an asynchronous context, and it is not +* allowed to use request_module() to load rc-cec.ko in that case. +* +* Since this keymap is only used if CONFIG_MEDIA_CEC_RC is set, we +* just compile this keymap into the rc-core module and never as a +* separate module. +* * Copyright (c) 2015 by Kamil Debski */
@@ -152,7 +162,7 @@ static struct rc_map_table cec[] = {/* 0x77-0xff: Reserved */ }; -static struct rc_map_list cec_map = { +struct rc_map_list cec_map = { .map = { .scan = cec, .size =ARRAY_SIZE (cec),
@@ -160,19 +170,3 @@ static struct rc_map_list cec_map = { .name = RC_MAP_CEC, } }; - -static int __init init_rc_map_cec (void ) -{ -return rc_map_register (&cec_map); -} - -static void __exit exit_rc_map_cec(void) -{ -rc_map_unregister (&cec_map); -} - -module_init (init_rc_map_cec); -module_exit (exit_rc_map_cec); - -MODULE_LICENSE ("GPL" ); -MODULE_AUTHOR ("Kamil Debski" );
drivers/media/rc/rc-main.c +6 -0
@@ -2033,6 +2033,9 @@ static int __init rc_core_init(void)led_trigger_register_simple ("rc-feedback" , &led_feedback);rc_map_register (&empty_map); +#ifdef CONFIG_MEDIA_CEC_RC +rc_map_register (&cec_map); +#endif return 0 ; }
@@ -2042,6 +2045,9 @@ static void __exit rc_core_exit(void)lirc_dev_exit ();class_unregister (&rc_class);led_trigger_unregister_simple (led_feedback); +#ifdef CONFIG_MEDIA_CEC_RC +rc_map_unregister (&cec_map); +#endif rc_map_unregister (&empty_map); }
include/media/rc-map.h +7 -0
@@ -126,6 +126,13 @@ struct rc_map_list {struct rc_map map; }; +#ifdef CONFIG_MEDIA_CEC_RC +/* +* rc_map_list from rc-cec.c +*/ +extern struct rc_map_list cec_map; +#endif +/* Routines from rc-map.c */ /**
[email protected]
maintainerPATCH 59/168 net: hns3: fix error mask definition of flow director
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Jian Shen <[email protected]>
[ Upstream commit ae85ddda0f1b341b2d25f5a5e0eff1d42b6ef3df ]
Currently, some bit filed definitions of flow director TCAM configuration command are incorrect. Since the wrong MSB is always 0, and these fields are assgined in order, so it still works.
Fix it by redefine them.
--- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h +3 -3
@@ -1018,16 +1018,16 @@ struct hclge_fd_tcam_config_3_cmd {#define HCLGE_FD_AD_DROP_B 0#define HCLGE_FD_AD_DIRECT_QID_B 1#define HCLGE_FD_AD_QID_S 2 -#define HCLGE_FD_AD_QID_M GENMASK(12, 2) +#define HCLGE_FD_AD_QID_M GENMASK(11, 2)#define HCLGE_FD_AD_USE_COUNTER_B 12#define HCLGE_FD_AD_COUNTER_NUM_S 13#define HCLGE_FD_AD_COUNTER_NUM_M GENMASK(20, 13)#define HCLGE_FD_AD_NXT_STEP_B 20#define HCLGE_FD_AD_NXT_KEY_S 21 -#define HCLGE_FD_AD_NXT_KEY_M GENMASK(26, 21) +#define HCLGE_FD_AD_NXT_KEY_M GENMASK(25, 21)#define HCLGE_FD_AD_WR_RULE_ID_B 0#define HCLGE_FD_AD_RULE_ID_S 1 -#define HCLGE_FD_AD_RULE_ID_M GENMASK(13, 1) +#define HCLGE_FD_AD_RULE_ID_M GENMASK(12, 1)struct hclge_fd_ad_config_cmd {u8 stage;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 60/168 net: enetc: initialize RFS/RSS memories for unused ports too
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Vladimir Oltean <[email protected]>
[ Upstream commit 3222b5b613db558e9a494bbf53f3c984d90f71ea ]
Michael reports that since linux-next-20210211, the AER messages for ECC errors have started reappearing, and this time they can be reliably reproduced with the first ping on one of his LS1028A boards.
$ ping 1[ 33.258069] pcieport 0000:00:1f.0: AER: Multiple Corrected error received: 0000:00:00.0 72.16.0.1 PING [ 33.267050] pcieport 0000:00:1f.0: AER: can't find device of ID0000 172.16.0.1 (172.16.0.1): 56 data bytes 64 bytes from 172.16.0.1: seq=0 ttl=64 time=17.124 ms 64 bytes from 172.16.0.1: seq=1 ttl=64 time=0.273 ms
$ devmem 0x1f8010e10 32 0xC0000006
It isn't clear why this is necessary, but it seems that for the errors to go away, we must clear the entire RFS and RSS memory, not just for the ports in use.
Sadly the code is structured in such a way that we can't have unified logic for the used and unused ports. For the minimal initialization of an unused port, we need just to enable and ioremap the PF memory space, and a control buffer descriptor ring. Unused ports must then free the CBDR because the driver will exit, but used ports can not pick up from where that code path left, since the CBDR API does not reinitialize a ring when setting it up, so its producer and consumer indices are out of sync between the software and hardware state. So a separate enetc_init_unused_port function was created, and it gets called right after the PF memory space is enabled.
--- drivers/net/ethernet/freescale/enetc/enetc.c | 8 ++--- drivers/net/ethernet/freescale/enetc/enetc.h | 4 +++ .../net/ethernet/freescale/enetc/enetc_pf.c | 33 ++++++++++++++++--- 3 files changed, 36 insertions(+), 9 deletions(-)
drivers/net/ethernet/freescale/enetc/enetc.c +4 -4
@@ -942,7 +942,7 @@ static void enetc_free_rxtx_rings(struct enetc_ndev_priv *priv)enetc_free_tx_ring (priv->tx_ring[i]); } -static int enetc_alloc_cbdr (struct device *dev,struct enetc_cbdr *cbdr) +int enetc_alloc_cbdr (struct device *dev,struct enetc_cbdr *cbdr) {int size = cbdr->bd_count *sizeof (struct enetc_cbd );
@@ -963,7 +963,7 @@ static int enetc_alloc_cbdr(struct device *dev, struct enetc_cbdr *cbdr)return 0 ; } -static void enetc_free_cbdr (struct device *dev,struct enetc_cbdr *cbdr) +void enetc_free_cbdr (struct device *dev,struct enetc_cbdr *cbdr) {int size = cbdr->bd_count *sizeof (struct enetc_cbd );
@@ -971,7 +971,7 @@ static void enetc_free_cbdr(struct device *dev, struct enetc_cbdr *cbdr) cbdr->bd_base =NULL ; } -static void enetc_setup_cbdr (struct enetc_hw *hw,struct enetc_cbdr *cbdr) +void enetc_setup_cbdr (struct enetc_hw *hw,struct enetc_cbdr *cbdr) {/* set CBDR cache attributes */ enetc_wr(hw, ENETC_SICAR2,
@@ -991,7 +991,7 @@ static void enetc_setup_cbdr(struct enetc_hw *hw, struct enetc_cbdr *cbdr) cbdr->cir = hw->reg +ENETC_SICBDRCIR ; } -static void enetc_clear_cbdr (struct enetc_hw *hw) +void enetc_clear_cbdr (struct enetc_hw *hw) {enetc_wr (hw,ENETC_SICBDRMR ,0 ); }
drivers/net/ethernet/freescale/enetc/enetc.h +4 -0
@@ -237,6 +237,10 @@ int enetc_setup_tc(struct net_device *ndev, enum tc_setup_type type,void enetc_set_ethtool_ops (struct net_device *ndev);/* control buffer descriptor ring (CBDR) */ +int enetc_alloc_cbdr (struct device *dev,struct enetc_cbdr *cbdr); +void enetc_free_cbdr (struct device *dev,struct enetc_cbdr *cbdr); +void enetc_setup_cbdr (struct enetc_hw *hw,struct enetc_cbdr *cbdr); +void enetc_clear_cbdr (struct enetc_hw *hw);int enetc_set_mac_flt_entry (struct enetc_si *si,int index,char *mac_addr,int si_map);int enetc_clear_mac_flt_entry (struct enetc_si *si,int index);
drivers/net/ethernet/freescale/enetc/enetc_pf.c +28 -5
@@ -854,6 +854,26 @@ static int enetc_init_port_rss_memory(struct enetc_si *si)return err; } +static void enetc_init_unused_port (struct enetc_si *si) +{ +struct device *dev = &si->pdev->dev; +struct enetc_hw *hw = &si->hw; +int err; + + si->cbd_ring.bd_count =ENETC_CBDR_DEFAULT_SIZE ; + err =enetc_alloc_cbdr (dev, &si->cbd_ring); +if (err) +return ; + +enetc_setup_cbdr (hw, &si->cbd_ring); + +enetc_init_port_rfs_memory (si); +enetc_init_port_rss_memory (si); + +enetc_clear_cbdr (hw); +enetc_free_cbdr (dev, &si->cbd_ring); +} +static int enetc_pf_probe (struct pci_dev *pdev,const struct pci_device_id *ent) {
@@ -863,11 +883,6 @@ static int enetc_pf_probe(struct pci_dev *pdev,struct enetc_pf *pf;int err; -if (pdev->dev.of_node && !of_device_is_available (pdev->dev.of_node)) { -dev_info (&pdev->dev,"device is disabled, skipping\n" ); -return -ENODEV ; - } - err =enetc_pci_probe (pdev,KBUILD_MODNAME ,sizeof (*pf));if (err) {dev_err (&pdev->dev,"PCI probing failed\n" );
@@ -881,6 +896,13 @@ static int enetc_pf_probe(struct pci_dev *pdev, goto err_map_pf_space; } +if (pdev->dev.of_node && !of_device_is_available (pdev->dev.of_node)) { +enetc_init_unused_port (si); +dev_info (&pdev->dev,"device is disabled, skipping\n" ); + err = -ENODEV ; + goto err_device_disabled; + } + pf =enetc_si_priv (si); pf->si = si; pf->total_vfs =pci_sriov_get_totalvfs (pdev);
@@ -960,6 +982,7 @@ err_alloc_si_res: si->ndev =NULL ;free_netdev (ndev); err_alloc_netdev: +err_device_disabled: err_map_pf_space:enetc_pci_remove (pdev);
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 61/168 net: phy: fix save wrong speed and duplex problem if autoneg is on
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Guangbin Huang <[email protected]>
[ Upstream commit d9032dba5a2b2bbf0fdce67c8795300ec9923b43 ]
If phy uses generic driver and autoneg is on, enter command "ethtool -s eth0 speed 50" will not change phy speed actually, but command "ethtool eth0" shows speed is 50Mb/s because phydev->speed has been set to 50 and no update later.
And duplex setting has same problem too.
However, if autoneg is on, phy only changes speed and duplex according to phydev->advertising, but not phydev->speed and phydev->duplex. So in this case, phydev->speed and phydev->duplex don't need to be set in function phy_ethtool_ksettings_set() if autoneg is on.
--- drivers/net/phy/phy.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/net/phy/phy.c +4 -3
@@ -345,15 +345,16 @@ int phy_ethtool_ksettings_set(struct phy_device *phydev, phydev->autoneg = autoneg; - phydev->speed = speed; +if (autoneg ==AUTONEG_DISABLE ) { + phydev->speed = speed; + phydev->duplex = duplex; + }linkmode_copy (phydev->advertising, advertising);linkmode_mod_bit (ETHTOOL_LINK_MODE_Autoneg_BIT, phydev->advertising, autoneg ==AUTONEG_ENABLE ); - phydev->duplex = duplex; - phydev->mdix_ctrl = cmd->base.eth_tp_mdix_ctrl;/* Restart the PHY */
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 62/168 i2c: rcar: faster irq code to minimize HW race condition
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Wolfram Sang <[email protected]>
[ Upstream commit c7b514ec979e23a08c411f3d8ed39c7922751422 ]
To avoid the HW race condition on R-Car Gen2 and earlier, we need to write to ICMCR as soon as possible in the interrupt handler. We can improve this by writing a static value instead of masking out bits.
--- drivers/i2c/busses/i2c-rcar.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-)
drivers/i2c/busses/i2c-rcar.c +4 -7
@@ -89,7 +89,6 @@#define RCAR_BUS_PHASE_START (MDBS | MIE | ESG)#define RCAR_BUS_PHASE_DATA (MDBS | MIE) -#define RCAR_BUS_MASK_DATA (~(ESG | FSB) & 0xFF)#define RCAR_BUS_PHASE_STOP (MDBS | MIE | FSB)#define RCAR_IRQ_SEND (MNR | MAL | MST | MAT | MDE)
@@ -616,7 +615,7 @@ static bool rcar_i2c_slave_irq(struct rcar_i2c_priv *priv)/* * This driver has a lock-free design because there are IP cores (at least * R-Car Gen2) which have an inherent race condition in their hardware design. -* There, we need to clear RCAR_BUS_MASK_DATA bits as soon as possible after +* There, we need to switch to RCAR_BUS_PHASE_DATA as soon as possible after * the interrupt was generated, otherwise an unwanted repeated message gets * generated. It turned out that taking a spinlock at the beginning of the ISR * was already causing repeated messages. Thus, this driver was converted to
@@ -625,13 +624,11 @@ static bool rcar_i2c_slave_irq(struct rcar_i2c_priv *priv)static irqreturn_t rcar_i2c_irq (int irq,void *ptr) {struct rcar_i2c_priv *priv = ptr; -u32 msr, val; +u32 msr;/* Clear START or STOP immediately, except for REPSTART after read */ -if (likely (!(priv->flags &ID_P_REP_AFTER_RD ))) { - val =rcar_i2c_read (priv,ICMCR ); -rcar_i2c_write (priv,ICMCR , val &RCAR_BUS_MASK_DATA ); - } +if (likely (!(priv->flags &ID_P_REP_AFTER_RD ))) +rcar_i2c_write (priv,ICMCR ,RCAR_BUS_PHASE_DATA ); msr =rcar_i2c_read (priv,ICMSR );
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 63/168 i2c: rcar: optimize cacheline to minimize HW race condition
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Wolfram Sang <[email protected]>
[ Upstream commit 25c2e0fb5fefb8d7847214cf114d94c7aad8e9ce ]
'flags' and 'io' are needed first, so they should be at the beginning of the private struct.
--- drivers/i2c/busses/i2c-rcar.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/i2c/busses/i2c-rcar.c +1 -1
@@ -116,6 +116,7 @@ enum rcar_i2c_type { };struct rcar_i2c_priv { +u32 flags;void __iomem *io;struct i2c_adapter adap;struct i2c_msg *msg;
@@ -126,7 +127,6 @@ struct rcar_i2c_priv {int pos;u32 icccr; -u32 flags;u8 recovery_icmcr;/* protected by adapter lock */ enum rcar_i2c_type devtype;struct i2c_client *slave;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 64/168 udf: fix silent AED tagLocation corruption
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Steven J. Magnani <[email protected]>
[ Upstream commit 63c9e47a1642fc817654a1bc18a6ec4bbcc0f056 ]
When extending a file, udf_do_extend_file() may enter following empty indirect extent. At the end of udf_do_extend_file() we revert prev_epos to point to the last written extent. However if we end up not adding any further extent in udf_do_extend_file(), the reverting points prev_epos into the header area of the AED and following updates of the extents (in udf_update_extents()) will corrupt the header.
Make sure that we do not follow indirect extent if we are not going to add any more extents so that returning back to the last written extent works correctly.
--- fs/udf/inode.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
fs/udf/inode.c +6 -3
@@ -547,11 +547,14 @@ static int udf_do_extend_file(struct inode *inode,udf_write_aext (inode, last_pos, &last_ext->extLocation, last_ext->extLength,1 ); +/* -* We've rewritten the last extent but there may be empty -* indirect extent after it - enter it. +* We've rewritten the last extent. If we are going to add +* more extents, we may need to enter possible following +* empty indirect extent. */ -udf_next_aext (inode, last_pos, &tmploc, &tmplen,0 ); +if (new_block_bytes || prealloc_len) +udf_next_aext (inode, last_pos, &tmploc, &tmplen,0 ); }/* Managed to do everything necessary? */
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 65/168 mmc: mxs-mmc: Fix a resource leak in an error handling path in mxs_mmc_probe()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Christophe JAILLET <[email protected]>
[ Upstream commit 0bb7e560f821c7770973a94e346654c4bdccd42c ]
If 'mmc_of_parse()' fails, we must undo the previous 'dma_request_chan()' call.
--- drivers/mmc/host/mxs-mmc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mmc/host/mxs-mmc.c +1 -1
@@ -644,7 +644,7 @@ static int mxs_mmc_probe(struct platform_device *pdev) ret =mmc_of_parse (mmc);if (ret) - goto out_clk_disable; + goto out_free_dma; mmc->ocr_avail =MMC_VDD_32_33 |MMC_VDD_33_34 ;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 66/168 mmc: mediatek: fix race condition between msdc_request_timeout and irq
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Chaotian Jing <[email protected]>
[ Upstream commit 0354ca6edd464a2cf332f390581977b8699ed081 ]
when get request SW timeout, if CMD/DAT xfer done irq coming right now, then there is race between the msdc_request_timeout work and irq handler, and the host->cmd and host->data may set to NULL in irq handler. also, current flow ensure that only one path can go to msdc_request_done(), so no need check the return value of cancel_delayed_work().
--- drivers/mmc/host/mtk-sd.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)
drivers/mmc/host/mtk-sd.c +10 -8
@@ -1020,13 +1020,13 @@ static void msdc_track_cmd_data(struct msdc_host *host,static void msdc_request_done (struct msdc_host *host,struct mmc_request *mrq) {unsigned long flags; -bool ret; - ret =cancel_delayed_work (&host->req_timeout); -if (!ret) { -/* delay work already running */ -return ; - } +/* +* No need check the return value of cancel_delayed_work, as only ONE +* path will go here! +*/ +cancel_delayed_work (&host->req_timeout); +spin_lock_irqsave (&host->lock, flags); host->mrq =NULL ;spin_unlock_irqrestore (&host->lock, flags);
@@ -1046,7 +1046,7 @@ static bool msdc_cmd_done(struct msdc_host *host, int events,bool done = false;bool sbc_error;unsigned long flags; -u32 *rsp = cmd->resp; +u32 *rsp; if (mrq->sbc && cmd == mrq->cmd && (events & (MSDC_INT_ACMDRDY | MSDC_INT_ACMDCRCERR
@@ -1067,6 +1067,7 @@ static bool msdc_cmd_done(struct msdc_host *host, int events,if (done)return true; + rsp = cmd->resp;sdr_clr_bits (host->base +MSDC_INTEN , cmd_ints_mask);
@@ -1254,7 +1255,7 @@ static void msdc_data_xfer_next(struct msdc_host *host,static bool msdc_data_xfer_done (struct msdc_host *host,u32 events,struct mmc_request *mrq,struct mmc_data *data) { -struct mmc_command *stop = data->stop; +struct mmc_command *stop;unsigned long flags;bool done; unsigned int check_data = events &
@@ -1270,6 +1271,7 @@ static bool msdc_data_xfer_done(struct msdc_host *host, u32 events,if (done)return true; + stop = data->stop;if (check_data || (stop && stop->error)) { dev_dbg(host->dev, "DMA status: 0x%8X\n",
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 67/168 Platform: OLPC: Fix probe error handling
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Lubomir Rintel <[email protected]>
[ Upstream commit cec551ea0d41c679ed11d758e1a386e20285b29d ]
Reset ec_priv if probe ends unsuccessfully.
--- drivers/platform/olpc/olpc-ec.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-)
drivers/platform/olpc/olpc-ec.c +8 -7
@@ -426,11 +426,8 @@ static int olpc_ec_probe(struct platform_device *pdev)/* get the EC revision */ err =olpc_ec_cmd (EC_FIRMWARE_REV ,NULL ,0 , &ec->version,1 ); -if (err) { - ec_priv =NULL ; -kfree (ec); -return err; - } +if (err) + goto error; config.dev = pdev->dev.parent; config.driver_data = ec;
@@ -440,12 +437,16 @@ static int olpc_ec_probe(struct platform_device *pdev)if (IS_ERR (ec->dcon_rdev)) {dev_err (&pdev->dev,"failed to register DCON regulator\n" ); err =PTR_ERR (ec->dcon_rdev); -kfree (ec); -return err; + goto error; } ec->dbgfs_dir =olpc_ec_setup_debugfs (); +return 0 ; + +error: + ec_priv =NULL ; +kfree (ec);return err; }
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 68/168 powerpc/pci: Add ppc_md.discover_phbs()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Oliver O'Halloran <[email protected]>
[ Upstream commit 5537fcb319d016ce387f818dd774179bc03217f5 ]
On many powerpc platforms the discovery and initalisation of pci_controllers (PHBs) happens inside of setup_arch(). This is very early in boot (pre-initcalls) and means that we're initialising the PHB long before many basic kernel services (slab allocator, debugfs, a real ioremap) are available.
On PowerNV this causes an additional problem since we map the PHB registers with ioremap(). As of commit d538aadc2718 ("powerpc/ioremap: warn on early use of ioremap()") a warning is printed because we're using the "incorrect" API to setup and MMIO mapping in searly boot. The kernel does provide early_ioremap(), but that is not intended to create long-lived MMIO mappings and a seperate warning is printed by generic code if early_ioremap() mappings are "leaked."
This is all fixable with dumb hacks like using early_ioremap() to setup the initial mapping then replacing it with a real ioremap later on in boot, but it does raise the question: Why the hell are we setting up the PHB's this early in boot?
The old and wise claim it's due to "hysterical rasins." Aside from amused grapes there doesn't appear to be any real reason to maintain the current behaviour. Already most of the newer embedded platforms perform PHB discovery in an arch_initcall and between the end of setup_arch() and the start of initcalls none of the generic kernel code does anything PCI related. On powerpc scanning PHBs occurs in a subsys_initcall so it should be possible to move the PHB discovery to a core, postcore or arch initcall.
This patch adds the ppc_md.discover_phbs hook and a core_initcall stub that calls it. The core_initcalls are the earliest to be called so this will any possibly issues with dependency between initcalls. This isn't just an academic issue either since on pseries and PowerNV EEH init occurs in an arch_initcall and depends on the pci_controllers being available, similarly the creation of pci_dns occurs at core_initcall_sync (i.e. between core and postcore initcalls). These problems need to be addressed seperately.
--- arch/powerpc/include/asm/machdep.h | 3 +++ arch/powerpc/kernel/pci-common.c | 10 ++++++++++ 2 files changed, 13 insertions(+)
arch/powerpc/include/asm/machdep.h +3 -0
@@ -59,6 +59,9 @@ struct machdep_calls {int (*pcibios_root_bridge_prepare)(struct pci_host_bridge *bridge); +/* finds all the pci_controllers present at boot */ +void (*discover_phbs)(void ); +/* To setup PHBs when using automatic OF platform driver for PCI */ int (*pci_setup_phb)(struct pci_controller *host);
arch/powerpc/kernel/pci-common.c +10 -0
@@ -1669,3 +1669,13 @@ static void fixup_hide_host_resource_fsl(struct pci_dev *dev) }DECLARE_PCI_FIXUP_HEADER (PCI_VENDOR_ID_MOTOROLA ,PCI_ANY_ID , fixup_hide_host_resource_fsl);DECLARE_PCI_FIXUP_HEADER (PCI_VENDOR_ID_FREESCALE ,PCI_ANY_ID , fixup_hide_host_resource_fsl); + + +static int __init discover_phbs (void ) +{ +if (ppc_md.discover_phbs) + ppc_md.discover_phbs (); + +return 0 ; +} +core_initcall (discover_phbs);
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 69/168 spi: stm32: make spurious and overrun interrupts visible
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Alain Volmat <[email protected]>
[ Upstream commit c64e7efe46b7de21937ef4b3594d9b1fc74f07df ]
We do not expect to receive spurious interrupts so rise a warning if it happens.
RX overrun is an error condition that signals a corrupted RX stream both in dma and in irq modes. Report the error and abort the transfer in either cases.
--- drivers/spi/spi-stm32.c | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-)
drivers/spi/spi-stm32.c +4 -11
@@ -924,8 +924,8 @@ static irqreturn_t stm32h7_spi_irq_thread(int irq, void *dev_id) mask |=STM32H7_SPI_SR_RXP ;if (!(sr & mask)) { -dev_dbg (spi->dev,"spurious IT (sr=0x%08x, ier=0x%08x)\n" , - sr, ier); +dev_warn (spi->dev,"spurious IT (sr=0x%08x, ier=0x%08x)\n" , + sr, ier);spin_unlock_irqrestore (&spi->lock, flags);return IRQ_NONE ; }
@@ -952,15 +952,8 @@ static irqreturn_t stm32h7_spi_irq_thread(int irq, void *dev_id) }if (sr &STM32H7_SPI_SR_OVR ) { -dev_warn (spi->dev,"Overrun: received value discarded\n" ); -if (!spi->cur_usedma && (spi->rx_buf && (spi->rx_len >0 ))) -stm32h7_spi_read_rxfifo (spi, false); -/* -* If overrun is detected while using DMA, it means that -* something went wrong, so stop the current transfer -*/ -if (spi->cur_usedma) - end = true; +dev_err (spi->dev,"Overrun: RX data lost\n" ); + end = true; }if (sr &STM32H7_SPI_SR_EOT ) {
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 70/168 powerpc: improve handling of unrecoverable system reset
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Nicholas Piggin <[email protected]>
[ Upstream commit 11cb0a25f71818ca7ab4856548ecfd83c169aa4d ]
If an unrecoverable system reset hits in process context, the system does not have to panic. Similar to machine check, call nmi_exit() before die().
--- arch/powerpc/kernel/traps.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
arch/powerpc/kernel/traps.c +4 -1
@@ -513,8 +513,11 @@ void system_reset_exception(struct pt_regs *regs)die ("Unrecoverable nested System Reset" , regs,SIGABRT );#endif /* Must die if the interrupt is not recoverable */ -if (!(regs->msr &MSR_RI )) +if (!(regs->msr &MSR_RI )) { +/* For the reason explained in die_mce, nmi_exit before die */ +nmi_exit ();die ("Unrecoverable System Reset" , regs,SIGABRT ); + }if (saved_hsrrs) {mtspr (SPRN_HSRR0 , hsrr0);
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 71/168 powerpc/perf: Record counter overflow always if SAMPLE_IP is unset
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Athira Rajeev <[email protected]>
[ Upstream commit d137845c973147a22622cc76c7b0bc16f6206323 ]
While sampling for marked events, currently we record the sample only if the SIAR valid bit of Sampled Instruction Event Register (SIER) is set. SIAR_VALID bit is used for fetching the instruction address from Sampled Instruction Address Register(SIAR). But there are some usecases, where the user is interested only in the PMU stats at each counter overflow and the exact IP of the overflow event is not required. Dropping SIAR invalid samples will fail to record some of the counter overflows in such cases.
Example of such usecase is dumping the PMU stats (event counts) after some regular amount of instructions/events from the userspace (ex: via ptrace). Here counter overflow is indicated to userspace via signal handler, and captured by monitoring and enabling I/O signaling on the event file descriptor. In these cases, we expect to get sample/overflow indication after each specified sample_period.
Perf event attribute will not have PERF_SAMPLE_IP set in the sample_type if exact IP of the overflow event is not requested. So while profiling if SAMPLE_IP is not set, just record the counter overflow irrespective of SIAR_VALID check.
--- arch/powerpc/perf/core-book3s.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-)
arch/powerpc/perf/core-book3s.c +15 -4
@@ -2077,7 +2077,17 @@ static void record_and_restart(struct perf_event *event, unsigned long val, left += period;if (left <=0 ) left = period; - record =siar_valid (regs); + +/* +* If address is not requested in the sample via +* PERF_SAMPLE_IP, just record that sample irrespective +* of SIAR valid check. +*/ +if (event->attr.sample_type &PERF_SAMPLE_IP ) + record =siar_valid (regs); +else + record =1 ; + event->hw.last_period = event->hw.sample_period; }if (left <0x80000000LL )
@@ -2095,9 +2105,10 @@ static void record_and_restart(struct perf_event *event, unsigned long val,* MMCR2. Check attr.exclude_kernel and address to drop the sample in * these cases. */ -if (event->attr.exclude_kernel && record) -if (is_kernel_addr (mfspr (SPRN_SIAR ))) - record =0 ; +if (event->attr.exclude_kernel && + (event->attr.sample_type &PERF_SAMPLE_IP ) && +is_kernel_addr (mfspr (SPRN_SIAR ))) + record =0 ;/* * Finally record data if requested.
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 72/168 HID: logitech-dj: add support for the new lightspeed connection iteration
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Filipe Laíns <[email protected]>
[ Upstream commit fab3a95654eea01d6b0204995be8b7492a00d001 ]
This new connection type is the new iteration of the Lightspeed connection and will probably be used in some of the newer gaming devices. It is currently use in the G Pro X Superlight.
This patch should be backported to older versions, as currently the driver will panic when seing the unsupported connection. This isn't an issue when using the receiver that came with the device, as Logitech has been using different PIDs when they change the connection type, but is an issue when using a generic receiver (well, generic Lightspeed receiver), which is the case of the one in the Powerplay mat. Currently, the only generic Ligthspeed receiver we support, and the only one that exists AFAIK, is ther Powerplay.
As it stands, the driver will panic when seeing a G Pro X Superlight connected to the Powerplay receiver and won't send any input events to userspace! The kernel will warn about this so the issue should be easy to identify, but it is still very worrying how hard it will fail :(
[915977.398471] logitech-djreceiver 0003:046D:C53A.0107: unusable device of type UNKNOWN (0x0f) connected on slot 1
--- drivers/hid/hid-logitech-dj.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/hid/hid-logitech-dj.c +6 -1
@@ -995,7 +995,12 @@ static void logi_hidpp_recv_queue_notif(struct hid_device *hdev, workitem.reports_supported |=STD_KEYBOARD ;break ;case 0x0d : - device_type ="eQUAD Lightspeed 1_1" ; + device_type ="eQUAD Lightspeed 1.1" ; +logi_hidpp_dev_conn_notif_equad (hdev, hidpp_report, &workitem); + workitem.reports_supported |=STD_KEYBOARD ; +break ; +case 0x0f : + device_type ="eQUAD Lightspeed 1.2" ;logi_hidpp_dev_conn_notif_equad (hdev, hidpp_report, &workitem); workitem.reports_supported |=STD_KEYBOARD ;break ;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 73/168 powerpc/64: Fix stack trace not displaying final frame
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Michael Ellerman <[email protected]>
[ Upstream commit e3de1e291fa58a1ab0f471a4b458eff2514e4b5f ]
In commit bf13718bc57a ("powerpc: show registers when unwinding interrupt frames") we changed our stack dumping logic to show the full registers whenever we find an interrupt frame on the stack.
However we didn't notice that on 64-bit this doesn't show the final frame, ie. the interrupt that brought us in from userspace, whereas on 32-bit it does.
That is due to confusion about the size of that last frame. The code in show_stack() calls validate_sp(), passing it STACK_INT_FRAME_SIZE to check the sp is at least that far below the top of the stack.
However on 64-bit that size is too large for the final frame, because it includes the red zone, but we don't allocate a red zone for the first frame.
So add a new define that encodes the correct size for 32-bit and 64-bit, and use it in show_stack().
This results in the full trace being shown on 64-bit, eg:
sysrq: Trigger a crash
Kernel panic - not syncing: sysrq triggered crash CPU: 0 PID: 83 Comm: sh Not tainted 5.11.0-rc2-gcc-8.2.0-00188-g571abcb96b10-dirty #649 Call Trace: [c00000000a1c3ac0] [c000000000897b70] dump_stack+0xc4/0x114 (unreliable) [c00000000a1c3b00] [c00000000014334c] panic+0x178/0x41c [c00000000a1c3ba0] [c00000000094e600] sysrq_handle_crash+0x40/0x50 [c00000000a1c3c00] [c00000000094ef98] __handle_sysrq+0xd8/0x210 [c00000000a1c3ca0] [c00000000094f820] write_sysrq_trigger+0x100/0x188 [c00000000a1c3ce0] [c0000000005559dc] proc_reg_write+0x10c/0x1b0 [c00000000a1c3d10] [c000000000479950] vfs_write+0xf0/0x360 [c00000000a1c3d60] [c000000000479d9c] ksys_write+0x7c/0x140 [c00000000a1c3db0] [c00000000002bf5c] system_call_exception+0x19c/0x2c0 [c00000000a1c3e10] [c00000000000d35c] system_call_common+0xec/0x278 --- interrupt: c00 at 0x7fff9fbab428 NIP: 00007fff9fbab428 LR: 000000001000b724 CTR: 0000000000000000 REGS: c00000000a1c3e80 TRAP: 0c00 Not tainted (5.11.0-rc2-gcc-8.2.0-00188-g571abcb96b10-dirty) MSR: 900000000280f033 <SF,HV,VEC,VSX,EE,PR,FP,ME,IR,DR,RI,LE> CR: 22002884 XER: 00000000 IRQMASK: 0 GPR00: 0000000000000004 00007fffc3cb8960 00007fff9fc59900 0000000000000001 GPR04: 000000002a4b32d0 0000000000000002 0000000000000063 0000000000000063 GPR08: 000000002a4b32d0 0000000000000000 0000000000000000 0000000000000000 GPR12: 0000000000000000 00007fff9fcca9a0 0000000000000000 0000000000000000 GPR16: 0000000000000000 0000000000000000 0000000000000000 00000000100b8fd0 GPR20: 000000002a4b3485 00000000100b8f90 0000000000000000 0000000000000000 GPR24: 000000002a4b0440 00000000100e77b8 0000000000000020 000000002a4b32d0 GPR28: 0000000000000001 0000000000000002 000000002a4b32d0 0000000000000001 NIP [00007fff9fbab428] 0x7fff9fbab428 LR [000000001000b724] 0x1000b724 --- interrupt: c00
--- arch/powerpc/include/asm/ptrace.h | 3 +++ arch/powerpc/kernel/asm-offsets.c | 2 +- arch/powerpc/kernel/process.c | 2 +- 3 files changed, 5 insertions(+), 2 deletions(-)
arch/powerpc/include/asm/ptrace.h +3 -0
@@ -62,6 +62,9 @@ struct pt_regs };#endif + +#define STACK_FRAME_WITH_PT_REGS (STACK_FRAME_OVERHEAD + sizeof(struct pt_regs)) +#ifdef __powerpc64__/*
arch/powerpc/kernel/asm-offsets.c +1 -1
@@ -285,7 +285,7 @@ int main(void)/* Interrupt register frame */ DEFINE (INT_FRAME_SIZE ,STACK_INT_FRAME_SIZE ); -DEFINE (SWITCH_FRAME_SIZE ,STACK_FRAME_OVERHEAD +sizeof (struct pt_regs )); +DEFINE (SWITCH_FRAME_SIZE ,STACK_FRAME_WITH_PT_REGS );STACK_PT_REGS_OFFSET (GPR0 , gpr[0 ]);STACK_PT_REGS_OFFSET (GPR1 , gpr[1 ]);STACK_PT_REGS_OFFSET (GPR2 , gpr[2 ]);
arch/powerpc/kernel/process.c +1 -1
@@ -2081,7 +2081,7 @@ void show_stack(struct task_struct *tsk, unsigned long *stack)* See if this is an exception frame. * We look for the "regshere" marker in the current frame. */ -if (validate_sp (sp, tsk,STACK_INT_FRAME_SIZE ) +if (validate_sp (sp, tsk,STACK_FRAME_WITH_PT_REGS ) && stack[STACK_FRAME_MARKER ] ==STACK_FRAME_REGS_MARKER ) {struct pt_regs *regs = (struct pt_regs *) (sp +STACK_FRAME_OVERHEAD );
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 74/168 iommu/amd: Fix performance counter initialization
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Suravee Suthikulpanit <[email protected]>
[ Upstream commit 6778ff5b21bd8e78c8bd547fd66437cf2657fd9b ]
Certain AMD platforms enable power gating feature for IOMMU PMC, which prevents the IOMMU driver from updating the counter while trying to validate the PMC functionality in the init_iommu_perf_ctr(). This results in disabling PMC support and the following error message:
"AMD-Vi: Unable to read/write to IOMMU perf counter"
To workaround this issue, disable power gating temporarily by programming the counter source to non-zero value while validating the counter, and restore the prior state afterward.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=201753
--- drivers/iommu/amd_iommu_init.c | 45 +++++++++++++++++++++++++--------- 1 file changed, 34 insertions(+), 11 deletions(-)
drivers/iommu/amd_iommu_init.c +34 -11
@@ -12,6 +12,7 @@#include <linux/acpi.h> #include <linux/list.h> #include <linux/bitmap.h> +#include <linux/delay.h> #include <linux/slab.h> #include <linux/syscore_ops.h> #include <linux/interrupt.h>
@@ -253,6 +254,8 @@ static enum iommu_init_state init_state = IOMMU_START_STATE;static int amd_iommu_enable_interrupts (void );static int __init iommu_go_to_state (enum iommu_init_state state);static void init_device_table_dma (void ); +static int iommu_pc_get_set_reg (struct amd_iommu *iommu,u8 bank,u8 cntr, +u8 fxn,u64 *value,bool is_write);static bool amd_iommu_pre_enabled = true;
@@ -1672,13 +1675,11 @@ static int __init init_iommu_all(struct acpi_table_header *table)return 0 ; } -static int iommu_pc_get_set_reg (struct amd_iommu *iommu,u8 bank,u8 cntr, -u8 fxn,u64 *value,bool is_write); - -static void init_iommu_perf_ctr (struct amd_iommu *iommu) +static void __init init_iommu_perf_ctr (struct amd_iommu *iommu) { +int retry;struct pci_dev *pdev = iommu->dev; -u64 val =0xabcd , val2 =0 , save_reg =0 ; +u64 val =0xabcd , val2 =0 , save_reg, save_src;if (!iommu_feature (iommu,FEATURE_PC ))return ;
@@ -1686,17 +1687,39 @@ static void init_iommu_perf_ctr(struct amd_iommu *iommu) amd_iommu_pc_present = true;/* save the value to restore, if writable */ -if (iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &save_reg, false)) +if (iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &save_reg, false) || +iommu_pc_get_set_reg (iommu,0 ,0 ,8 , &save_src, false)) goto pc_false; -/* Check if the performance counters can be written to */ -if ((iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &val, true)) || - (iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &val2, false)) || - (val != val2)) +/* +* Disable power gating by programing the performance counter +* source to 20 (i.e. counts the reads and writes from/to IOMMU +* Reserved Register [MMIO Offset 1FF8h] that are ignored.), +* which never get incremented during this init phase. +* (Note: The event is also deprecated.) +*/ + val =20 ; +if (iommu_pc_get_set_reg (iommu,0 ,0 ,8 , &val, true)) goto pc_false; +/* Check if the performance counters can be written to */ + val =0xabcd ; +for (retry =5 ; retry; retry--) { +if (iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &val, true) || +iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &val2, false) || + val2) +break ; + +/* Wait about 20 msec for power gating to disable and retry. */ +msleep (20 ); + } +/* restore */ -if (iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &save_reg, true)) +if (iommu_pc_get_set_reg (iommu,0 ,0 ,0 , &save_reg, true) || +iommu_pc_get_set_reg (iommu,0 ,0 ,8 , &save_src, true)) + goto pc_false; + +if (val != val2) goto pc_false;pci_info (pdev,"IOMMU performance counters supported\n" );
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 75/168 sparc32: Limit memblock allocation to low memory
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Andreas Larsson <[email protected]>
[ Upstream commit bda166930c37604ffa93f2425426af6921ec575a ]
Commit cca079ef8ac29a7c02192d2bad2ffe4c0c5ffdd0 changed sparc32 to use memblocks instead of bootmem, but also made high memory available via memblock allocation which does not work together with e.g. phys_to_virt and can lead to kernel panic.
This changes back to only low memory being allocatable in the early stages, now using memblock allocation.
--- arch/sparc/mm/init_32.c | 3 +++ 1 file changed, 3 insertions(+)
arch/sparc/mm/init_32.c +3 -0
@@ -197,6 +197,9 @@ unsigned long __init bootmem_init(unsigned long *pages_avail) size =memblock_phys_mem_size () -memblock_reserved_size (); *pages_avail = (size >>PAGE_SHIFT ) - high_pages; +/* Only allow low memory to be allocated via memblock allocation */ +memblock_set_current_limit (max_low_pfn <<PAGE_SHIFT ); +return max_pfn; }
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 76/168 sparc64: Use arch_validate_flags() to validate ADI flag
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Khalid Aziz <[email protected]>
[ Upstream commit 147d8622f2a26ef34beacc60e1ed8b66c2fa457f ]
When userspace calls mprotect() to enable ADI on an address range, do_mprotect_pkey() calls arch_validate_prot() to validate new protection flags. arch_validate_prot() for sparc looks at the first VMA associated with address range to verify if ADI can indeed be enabled on this address range. This has two issues - (1) Address range might cover multiple VMAs while arch_validate_prot() looks at only the first VMA, (2) arch_validate_prot() peeks at VMA without holding mmap lock which can result in race condition.
arch_validate_flags() from commit c462ac288f2c ("mm: Introduce arch_validate_flags()") allows for VMA flags to be validated for all VMAs that cover the address range given by user while holding mmap lock. This patch updates sparc code to move the VMA check from arch_validate_prot() to arch_validate_flags() to fix above two issues.
--- arch/sparc/include/asm/mman.h | 54 +++++++++++++++++++---------------- 1 file changed, 29 insertions(+), 25 deletions(-)
arch/sparc/include/asm/mman.h +29 -25
@@ -57,35 +57,39 @@ static inline int sparc_validate_prot(unsigned long prot, unsigned long addr) {if (prot & ~(PROT_READ |PROT_WRITE |PROT_EXEC |PROT_SEM |PROT_ADI ))return 0 ; -if (prot &PROT_ADI ) { -if (!adi_capable ()) -return 0 ; +return 1 ; +} -if (addr) { -struct vm_area_struct *vma; +#define arch_validate_flags (vm_flags) arch_validate_flags(vm_flags) +/* arch_validate_flags() - Ensure combination of flags is valid for a +* VMA. +*/ +static inline bool arch_validate_flags (unsigned long vm_flags) +{ +/* If ADI is being enabled on this VMA, check for ADI +* capability on the platform and ensure VMA is suitable +* for ADI +*/ +if (vm_flags &VM_SPARC_ADI ) { +if (!adi_capable ()) +return false; - vma =find_vma (current->mm, addr); -if (vma) { -/* ADI can not be enabled on PFN -* mapped pages -*/ -if (vma->vm_flags & (VM_PFNMAP |VM_MIXEDMAP )) -return 0 ; +/* ADI can not be enabled on PFN mapped pages */ +if (vm_flags & (VM_PFNMAP |VM_MIXEDMAP )) +return false; -/* Mergeable pages can become unmergeable -* if ADI is enabled on them even if they -* have identical data on them. This can be -* because ADI enabled pages with identical -* data may still not have identical ADI -* tags on them. Disallow ADI on mergeable -* pages. -*/ -if (vma->vm_flags &VM_MERGEABLE ) -return 0 ; - } - } +/* Mergeable pages can become unmergeable +* if ADI is enabled on them even if they +* have identical data on them. This can be +* because ADI enabled pages with identical +* data may still not have identical ADI +* tags on them. Disallow ADI on mergeable +* pages. +*/ +if (vm_flags &VM_MERGEABLE ) +return false; } -return 1 ; +return true; }#endif /* CONFIG_SPARC64 */
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 77/168 Input: applespi - dont wait for responses to commands indefinitely.
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ronald Tschalär <[email protected]>
[ Upstream commit 0ce1ac23149c6da939a5926c098c270c58c317a0 ]
The response to a command may never arrive or it may be corrupted (and hence dropped) for some reason. While exceedingly rare, when it did happen it blocked all further commands. One way to fix this was to do a suspend/resume. However, recovering automatically seems like a nicer option. Hence this puts a time limit (1 sec) on how long we're willing to wait for a response, after which we assume it got lost.
--- drivers/input/keyboard/applespi.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-)
drivers/input/keyboard/applespi.c +15 -6
@@ -48,6 +48,7 @@#include <linux/efi.h> #include <linux/input.h> #include <linux/input/mt.h> +#include <linux/ktime.h> #include <linux/leds.h> #include <linux/module.h> #include <linux/spinlock.h>
@@ -400,7 +401,7 @@ struct applespi_data {unsigned int cmd_msg_cntr;/* lock to protect the above parameters and flags below */ spinlock_t cmd_msg_lock; -bool cmd_msg_queued; +ktime_t cmd_msg_queued;enum applespi_evt_type cmd_evt_type;struct led_classdev backlight_info;
@@ -716,7 +717,7 @@ static void applespi_msg_complete(struct applespi_data *applespi,wake_up_all (&applespi->drain_complete);if (is_write_msg) { - applespi->cmd_msg_queued = false; + applespi->cmd_msg_queued =0 ;applespi_send_cmd_msg (applespi); }
@@ -758,8 +759,16 @@ static int applespi_send_cmd_msg(struct applespi_data *applespi)return 0 ;/* check whether send is in progress */ -if (applespi->cmd_msg_queued) -return 0 ; +if (applespi->cmd_msg_queued) { +if (ktime_ms_delta (ktime_get (), applespi->cmd_msg_queued) <1000 ) +return 0 ; + +dev_warn (&applespi->spi->dev,"Command %d timed out\n" , + applespi->cmd_evt_type); + + applespi->cmd_msg_queued =0 ; + applespi->write_active = false; + }/* set up packet */ memset (packet,0 ,APPLESPI_PACKET_SIZE );
@@ -856,7 +865,7 @@ static int applespi_send_cmd_msg(struct applespi_data *applespi)return sts; } - applespi->cmd_msg_queued = true; + applespi->cmd_msg_queued =ktime_get_coarse (); applespi->write_active = true;return 0 ;
@@ -1908,7 +1917,7 @@ static int __maybe_unused applespi_resume(struct device *dev) applespi->drain = false; applespi->have_cl_led_on = false; applespi->have_bl_level =0 ; - applespi->cmd_msg_queued = false; + applespi->cmd_msg_queued =0 ; applespi->read_active = false; applespi->write_active = false;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 78/168 PCI: xgene-msi: Fix race in installing chained irq handler
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Martin Kaiser <[email protected]>
[ Upstream commit a93c00e5f975f23592895b7e83f35de2d36b7633 ]
Fix a race where a pending interrupt could be received and the handler called before the handler's data has been setup, by converting to irq_set_chained_handler_and_data().
See also 2cf5a03cb29d ("PCI/keystone: Fix race in installing chained IRQ handler").
Based on the mail discussion, it seems ok to drop the error handling.
--- drivers/pci/controller/pci-xgene-msi.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-)
drivers/pci/controller/pci-xgene-msi.c +3 -7
@@ -384,13 +384,9 @@ static int xgene_msi_hwirq_alloc(unsigned int cpu)if (!msi_group->gic_irq)continue ; -irq_set_chained_handler (msi_group->gic_irq, - xgene_msi_isr); - err =irq_set_handler_data (msi_group->gic_irq, msi_group); -if (err) { -pr_err ("failed to register GIC IRQ handler\n" ); -return -EINVAL ; - } +irq_set_chained_handler_and_data (msi_group->gic_irq, + xgene_msi_isr, msi_group); +/* * Statically allocate MSI GIC IRQs to each CPU core. * With 8-core X-Gene v1, 2 MSI GIC IRQs are allocated
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 79/168 PCI: mediatek: Add missing of_node_put() to fix reference leak
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Krzysztof Wilczyński <[email protected]>
[ Upstream commit 42814c438aac79746d310f413a27d5b0b959c5de ]
The for_each_available_child_of_node helper internally makes use of the of_get_next_available_child() which performs an of_node_get() on each iteration when searching for next available child node.
Should an available child node be found, then it would return a device node pointer with reference count incremented, thus early return from the middle of the loop requires an explicit of_node_put() to prevent reference count leak.
To stop the reference leak, explicitly call of_node_put() before returning after an error occurred.
--- drivers/pci/controller/pcie-mediatek.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/pci/controller/pcie-mediatek.c +5 -2
@@ -1063,14 +1063,14 @@ static int mtk_pcie_setup(struct mtk_pcie *pcie) err =of_pci_get_devfn (child);if (err <0 ) {dev_err (dev,"failed to parse devfn: %d\n" , err); -return err; + goto error_put_node; } slot =PCI_SLOT (err); err =mtk_pcie_parse_port (pcie, child, slot);if (err) -return err; + goto error_put_node; } err =mtk_pcie_subsys_powerup (pcie);
@@ -1086,6 +1086,9 @@ static int mtk_pcie_setup(struct mtk_pcie *pcie)mtk_pcie_subsys_powerdown (pcie);return 0 ; +error_put_node: +of_node_put (child); +return err; }static int mtk_pcie_probe (struct platform_device *pdev)
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 80/168 kbuild: clamp SUBLEVEL to 255
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
[ Upstream commit 9b82f13e7ef316cdc0a8858f1349f4defce3f9e0 ]
Right now if SUBLEVEL becomes larger than 255 it will overflow into the territory of PATCHLEVEL, causing havoc in userspace that tests for specific kernel version.
While userspace code tests for MAJOR and PATCHLEVEL, it doesn't test SUBLEVEL at any point as ABI changes don't happen in the context of stable tree.
Thus, to avoid overflows, simply clamp SUBLEVEL to it's maximum value in the context of LINUX_VERSION_CODE. This does not affect "make kernelversion" and such.
--- Makefile | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
Makefile +9 -3
@@ -1175,9 +1175,15 @@ define filechk_utsrelease.h endef define filechk_version.h - echo \#define LINUX_VERSION_CODE $(shell \ - expr $(VERSION) \* 65536 + 0$(PATCHLEVEL) \* 256 + 0$(SUBLEVEL)); \ - echo '#define KERNEL_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + (c))' + if [ $(SUBLEVEL) -gt 255 ]; then \ + echo \#define LINUX_VERSION_CODE $(shell \ + expr $(VERSION) \* 65536 + 0$(PATCHLEVEL) \* 256 + 255); \ + else \ + echo \#define LINUX_VERSION_CODE $(shell \ + expr $(VERSION) \* 65536 + 0$(PATCHLEVEL) \* 256 + $(SUBLEVEL)); \ + fi; \ + echo '#define KERNEL_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + \ + ((c) > 255 ? 255 : (c)))' endef $(version_h): FORCE
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 81/168 PCI: Fix pci_register_io_range() memory leak
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Geert Uytterhoeven <[email protected]>
[ Upstream commit f6bda644fa3a7070621c3bf12cd657f69a42f170 ]
Kmemleak reports:
unreferenced object 0xc328de40 (size 64):
comm "kworker/1:1", pid 21, jiffies 4294938212 (age 1484.670s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 e0 d8 fc eb 00 00 00 00 ................
00 00 10 fe 00 00 00 00 00 00 00 00 00 00 00 00 ................backtrace:
[<ad758d10>] pci_register_io_range+0x3c/0x80
[<2c7f139e>] of_pci_range_to_resource+0x48/0xc0
[<f079ecc8>] devm_of_pci_get_host_bridge_resources.constprop.0+0x2ac/0x3ac
[<e999753b>] devm_of_pci_bridge_init+0x60/0x1b8
[<a895b229>] devm_pci_alloc_host_bridge+0x54/0x64
[<e451ddb0>] rcar_pcie_probe+0x2c/0x644In case a PCI host driver's probe is deferred, the same I/O range may be allocated again, and be ignored, causing a memory leak.
Fix this by (a) letting logic_pio_register_range() return -EEXIST if the passed range already exists, so pci_register_io_range() will free it, and by (b) making pci_register_io_range() not consider -EEXIST an error condition.
--- drivers/pci/pci.c | 4 ++++ lib/logic_pio.c | 3 +++ 2 files changed, 7 insertions(+)
drivers/pci/pci.c +4 -0
@@ -3903,6 +3903,10 @@ int pci_register_io_range(struct fwnode_handle *fwnode, phys_addr_t addr, ret =logic_pio_register_range (range);if (ret)kfree (range); + +/* Ignore duplicates due to deferred probing */ +if (ret == -EEXIST ) + ret =0 ;#endif return ret;
lib/logic_pio.c +3 -0
@@ -27,6 +27,8 @@ static DEFINE_MUTEX(io_range_mutex);* @new_range: pointer to the IO range to be registered. * * Returns 0 on success, the error code in case of failure. +* If the range already exists, -EEXIST will be returned, which should be +* considered a success. * * Register a new IO range node in the IO range list. */
@@ -49,6 +51,7 @@ int logic_pio_register_range(struct logic_pio_hwaddr *new_range)list_for_each_entry (range, &io_range_list, list) {if (range->fwnode == new_range->fwnode) {/* range already there */ + ret = -EEXIST ; goto end_register; } if (range->flags == LOGIC_PIO_CPU_MMIO &&
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 82/168 i40e: Fix memory leak in i40e_probe
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Keita Suzuki <[email protected]>
[ Upstream commit 58cab46c622d6324e47bd1c533693c94498e4172 ]
Struct i40e_veb is allocated in function i40e_setup_pf_switch, and stored to an array field veb inside struct i40e_pf. However when i40e_setup_misc_vector fails, this memory leaks.
Fix this by calling exit and teardown functions.
--- drivers/net/ethernet/intel/i40e/i40e_main.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/intel/i40e/i40e_main.c +2 -0
@@ -15142,6 +15142,8 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)if (err) {dev_info (&pdev->dev,"setup of misc vector failed: %d\n" , err); +i40e_cloud_filter_exit (pf); +i40e_fdir_teardown (pf); goto err_vsis; } }
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 83/168 s390/smp: __smp_rescan_cpus() - move cpumask away from stack
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Heiko Carstens <[email protected]>
[ Upstream commit 62c8dca9e194326802b43c60763f856d782b225c ]
Avoid a potentially large stack frame and overflow by making "cpumask_t avail" a static variable. There is no concurrent access due to the existing locking.
--- arch/s390/kernel/smp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/s390/kernel/smp.c +1 -1
@@ -765,7 +765,7 @@ static int smp_add_core(struct sclp_core_entry *core, cpumask_t *avail,static int __smp_rescan_cpus (struct sclp_core_info *info,bool early) {struct sclp_core_entry *core; -cpumask_t avail; +static cpumask_t avail;bool configured;u16 core_id;int nr, i;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 84/168 sysctl.c: fix underflow value setting risk in vm_table
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Lin Feng <[email protected]>
[ Upstream commit 3b3376f222e3ab58367d9dd405cafd09d5e37b7c ]
Apart from subsystem specific .proc_handler handler, all ctl_tables with extra1 and extra2 members set should use proc_dointvec_minmax instead of proc_dointvec, or the limit set in extra* never work and potentially echo underflow values(negative numbers) is likely make system unstable.
Especially vfs_cache_pressure and zone_reclaim_mode, -1 is apparently not a valid value, but we can set to them. And then kernel may crash.
# echo -1 > /proc/sys/vm/vfs_cache_pressure
--- kernel/sysctl.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
kernel/sysctl.c +4 -4
@@ -1563,7 +1563,7 @@ static struct ctl_table vm_table[] = { .data = &block_dump, .maxlen = sizeof(block_dump), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, {
@@ -1571,7 +1571,7 @@ static struct ctl_table vm_table[] = { .data = &sysctl_vfs_cache_pressure, .maxlen = sizeof(sysctl_vfs_cache_pressure), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, #if defined(HAVE_ARCH_PICK_MMAP_LAYOUT) || \
@@ -1581,7 +1581,7 @@ static struct ctl_table vm_table[] = { .data = &sysctl_legacy_va_layout, .maxlen = sizeof(sysctl_legacy_va_layout), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, },#endif
@@ -1591,7 +1591,7 @@ static struct ctl_table vm_table[] = { .data = &node_reclaim_mode, .maxlen = sizeof(node_reclaim_mode), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, {
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 85/168 scsi: libiscsi: Fix iscsi_prep_scsi_cmd_pdu() error handling
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Mike Christie <[email protected]>
[ Upstream commit d28d48c699779973ab9a3bd0e5acfa112bd4fdef ]
If iscsi_prep_scsi_cmd_pdu() fails we try to add it back to the cmdqueue, but we leave it partially setup. We don't have functions that can undo the pdu and init task setup. We only have cleanup_task which can clean up both parts. So this has us just fail the cmd and go through the standard cleanup routine and then have the SCSI midlayer retry it like is done when it fails in the queuecommand path.
--- drivers/scsi/libiscsi.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-)
drivers/scsi/libiscsi.c +3 -8
@@ -1532,14 +1532,9 @@ static int iscsi_data_xmit(struct iscsi_conn *conn) } rc =iscsi_prep_scsi_cmd_pdu (conn->task);if (rc) { -if (rc == -ENOMEM || rc == -EACCES ) { -spin_lock_bh (&conn->taskqueuelock); -list_add_tail (&conn->task->running, - &conn->cmdqueue); - conn->task =NULL ; -spin_unlock_bh (&conn->taskqueuelock); - goto done; - }else +if (rc == -ENOMEM || rc == -EACCES ) +fail_scsi_task (conn->task,DID_IMM_RETRY ); +else fail_scsi_task (conn->task,DID_ABORT );spin_lock_bh (&conn->taskqueuelock);continue ;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 86/168 scsi: target: core: Add cmd length set before cmd complete
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Aleksandr Miloserdov <[email protected]>
[ Upstream commit 1c73e0c5e54d5f7d77f422a10b03ebe61eaed5ad ]
TCM doesn't properly handle underflow case for service actions. One way to prevent it is to always complete command with target_complete_cmd_with_length(), however it requires access to data_sg, which is not always available.
This change introduces target_set_cmd_data_length() function which allows to set command data length before completing it.
--- drivers/target/target_core_transport.c | 15 +++++++++++---- include/target/target_core_backend.h | 1 + 2 files changed, 12 insertions(+), 4 deletions(-)
drivers/target/target_core_transport.c +11 -4
@@ -873,11 +873,9 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status) }EXPORT_SYMBOL (target_complete_cmd); -void target_complete_cmd_with_length (struct se_cmd *cmd,u8 scsi_status,int length) +void target_set_cmd_data_length (struct se_cmd *cmd,int length) { -if ((scsi_status ==SAM_STAT_GOOD || - cmd->se_cmd_flags &SCF_TREAT_READ_AS_NORMAL ) && - length < cmd->data_length) { +if (length < cmd->data_length) {if (cmd->se_cmd_flags &SCF_UNDERFLOW_BIT ) { cmd->residual_count += cmd->data_length - length; }else {
@@ -887,6 +885,15 @@ void target_complete_cmd_with_length(struct se_cmd *cmd, u8 scsi_status, int len cmd->data_length = length; } +} +EXPORT_SYMBOL (target_set_cmd_data_length); + +void target_complete_cmd_with_length (struct se_cmd *cmd,u8 scsi_status,int length) +{ +if (scsi_status ==SAM_STAT_GOOD || + cmd->se_cmd_flags &SCF_TREAT_READ_AS_NORMAL ) { +target_set_cmd_data_length (cmd, length); + }target_complete_cmd (cmd, scsi_status); }
include/target/target_core_backend.h +1 -0
@@ -69,6 +69,7 @@ int transport_backend_register(const struct target_backend_ops *);void target_backend_unregister (const struct target_backend_ops *);void target_complete_cmd (struct se_cmd *,u8 ); +void target_set_cmd_data_length (struct se_cmd *,int );void target_complete_cmd_with_length (struct se_cmd *,u8 ,int );void transport_copy_sense_to_cmd (struct se_cmd *,unsigned char *);
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 87/168 scsi: target: core: Prevent underflow for service actions
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Aleksandr Miloserdov <[email protected]>
[ Upstream commit 14d24e2cc77411301e906a8cf41884739de192de ]
TCM buffer length doesn't necessarily equal 8 + ADDITIONAL LENGTH which might be considered an underflow in case of Data-In size being greater than 8 + ADDITIONAL LENGTH. So truncate buffer length to prevent underflow.
--- drivers/target/target_core_pr.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-)
drivers/target/target_core_pr.c +11 -4
@@ -3731,6 +3731,7 @@ core_scsi3_pri_read_keys(struct se_cmd *cmd)spin_unlock (&dev->t10_pr.registration_lock);put_unaligned_be32 (add_len, &buf[4 ]); +target_set_cmd_data_length (cmd,8 + add_len);transport_kunmap_data_sg (cmd);
@@ -3749,7 +3750,7 @@ core_scsi3_pri_read_reservation(struct se_cmd *cmd)struct t10_pr_registration *pr_reg;unsigned char *buf;u64 pr_res_key; -u32 add_len =16 ;/* Hardcoded to 16 when a reservation is held. */ +u32 add_len =0 ;if (cmd->data_length <8 ) { pr_err("PRIN SA READ_RESERVATIONS SCSI Data Length: %u"
@@ -3767,8 +3768,9 @@ core_scsi3_pri_read_reservation(struct se_cmd *cmd) pr_reg = dev->dev_pr_res_holder;if (pr_reg) {/* -* Set the hardcoded Additional Length +* Set the Additional Length to 16 when a reservation is held */ + add_len =16 ;put_unaligned_be32 (add_len, &buf[4 ]);if (cmd->data_length <22 )
@@ -3804,6 +3806,8 @@ core_scsi3_pri_read_reservation(struct se_cmd *cmd) (pr_reg->pr_res_type &0x0f ); } +target_set_cmd_data_length (cmd,8 + add_len); + err:spin_unlock (&dev->dev_reservation_lock);transport_kunmap_data_sg (cmd);
@@ -3822,7 +3826,7 @@ core_scsi3_pri_report_capabilities(struct se_cmd *cmd)struct se_device *dev = cmd->se_dev;struct t10_reservation *pr_tmpl = &dev->t10_pr;unsigned char *buf; -u16 add_len =8 ;/* Hardcoded to 8. */ +u16 len =8 ;/* Hardcoded to 8. */ if (cmd->data_length <6 ) { pr_err("PRIN SA REPORT_CAPABILITIES SCSI Data Length:"
@@ -3834,7 +3838,7 @@ core_scsi3_pri_report_capabilities(struct se_cmd *cmd)if (!buf)return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE ; -put_unaligned_be16 (add_len, &buf[0 ]); +put_unaligned_be16 (len, &buf[0 ]); buf[2 ] |=0x10 ;/* CRH: Compatible Reservation Hanlding bit. */ buf[2 ] |=0x08 ;/* SIP_C: Specify Initiator Ports Capable bit */ buf[2 ] |=0x04 ;/* ATP_C: All Target Ports Capable bit */
@@ -3863,6 +3867,8 @@ core_scsi3_pri_report_capabilities(struct se_cmd *cmd) buf[4 ] |=0x02 ;/* PR_TYPE_WRITE_EXCLUSIVE */ buf[5 ] |=0x01 ;/* PR_TYPE_EXCLUSIVE_ACCESS_ALLREG */ +target_set_cmd_data_length (cmd, len); +transport_kunmap_data_sg (cmd);return 0 ;
@@ -4023,6 +4029,7 @@ core_scsi3_pri_read_full_status(struct se_cmd *cmd)* Set ADDITIONAL_LENGTH */ put_unaligned_be32 (add_len, &buf[4 ]); +target_set_cmd_data_length (cmd,8 + add_len);transport_kunmap_data_sg (cmd);
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 88/168 ALSA: usb: Add Plantronics C320-M USB ctrl msg delay quirk
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: John Ernberg <[email protected]>
commit fc7c5c208eb7bc2df3a9f4234f14eca250001cb6 upstream.
The microphone in the Plantronics C320-M headset will randomly fail to initialize properly, at least when using Microsoft Teams. Introducing a 20ms delay on the control messages appears to resolve the issue.
--- sound/usb/quirks.c | 8 ++++++++ 1 file changed, 8 insertions(+)
sound/usb/quirks.c +8 -0
@@ -1604,6 +1604,14 @@ void snd_usb_ctl_msg_quirk(struct usb_de && (requesttype & USB_TYPE_MASK) == USB_TYPE_CLASS)msleep (20 ); +/* +* Plantronics C320-M needs a delay to avoid random +* microhpone failures. +*/ +if (chip->usb_id ==USB_ID (0x047f ,0xc025 ) && + (requesttype &USB_TYPE_MASK ) ==USB_TYPE_CLASS ) +msleep (20 ); +/* Zoom R16/24, many Logitech(at least H650e/H570e/BCC950), * Jabra 550a, Kingston HyperX needs a tiny delay here, * otherwise requests like get/set frequency return
[email protected]
maintainerPATCH 89/168 ALSA: hda/hdmi: Cancel pending works before suspend
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Takashi Iwai <[email protected]>
commit eea46a0879bcca23e15071f9968c0f6e6596e470 upstream.
The per_pin->work might be still floating at the suspend, and this may hit the access to the hardware at an unexpected timing. Cancel the work properly at the suspend callback for avoiding the buggy access.
Note that the bug doesn't trigger easily in the recent kernels since the work is queued only when the repoll count is set, and usually it's only at the resume callback, but it's still possible to hit in theory.
--- sound/pci/hda/patch_hdmi.c | 13 +++++++++++++ 1 file changed, 13 insertions(+)
sound/pci/hda/patch_hdmi.c +13 -0
@@ -2382,6 +2382,18 @@ static void generic_hdmi_free(struct hda }#ifdef CONFIG_PM +static int generic_hdmi_suspend (struct hda_codec *codec) +{ +struct hdmi_spec *spec = codec->spec; +int pin_idx; + +for (pin_idx =0 ; pin_idx < spec->num_pins; pin_idx++) { +struct hdmi_spec_per_pin *per_pin =get_pin (spec, pin_idx); +cancel_delayed_work_sync (&per_pin->work); + } +return 0 ; +} +static int generic_hdmi_resume (struct hda_codec *codec) {struct hdmi_spec *spec = codec->spec;
@@ -2405,6 +2417,7 @@ static const struct hda_codec_ops generi .build_controls = generic_hdmi_build_controls, .unsol_event = hdmi_unsol_event, #ifdef CONFIG_PM + .suspend = generic_hdmi_suspend, .resume = generic_hdmi_resume,#endif };
[email protected]
maintainerPATCH 90/168 ALSA: hda/ca0132: Add Sound BlasterX AE-5 Plus support
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Simeon Simeonoff <[email protected]>
commit f15c5c11abfbf8909eb30598315ecbec2311cfdc upstream.
The new AE-5 Plus model has a different Subsystem ID compared to the non-plus model. Adding the new id to the list of quirks.
--- sound/pci/hda/patch_ca0132.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_ca0132.c +1 -0
@@ -1185,6 +1185,7 @@ static const struct snd_pci_quirk ca0132SND_PCI_QUIRK (0x1102 ,0x0013 ,"Recon3D" ,QUIRK_R3D ),SND_PCI_QUIRK (0x1102 ,0x0018 ,"Recon3D" ,QUIRK_R3D ),SND_PCI_QUIRK (0x1102 ,0x0051 ,"Sound Blaster AE-5" ,QUIRK_AE5 ), +SND_PCI_QUIRK (0x1102 ,0x0191 ,"Sound Blaster AE-5 Plus" ,QUIRK_AE5 ), SND_PCI_QUIRK(0x1102, 0x0081, "Sound Blaster AE-7", QUIRK_AE7), {} };
[email protected]
maintainerPATCH 91/168 ALSA: hda: Drop the BATCH workaround for AMD controllers
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Takashi Iwai <[email protected]>
commit 28e96c1693ec1cdc963807611f8b5ad400431e82 upstream.
The commit c02f77d32d2c ("ALSA: hda - Workaround for crackled sound on AMD controller (1022:1457)") introduced a few workarounds for the recent AMD HD-audio controller, and one of them is the forced BATCH PCM mode so that PulseAudio avoids the timer-based scheduling. This was thought to cover for some badly working applications, but this actually worsens for more others. In total, this wasn't a good idea to enforce it.
This is a partial revert of the commit above for dropping the PCM BATCH enforcement part to recover from the regression again.
--- sound/pci/hda/hda_controller.c | 7 ------- 1 file changed, 7 deletions(-)
sound/pci/hda/hda_controller.c +0 -7
@@ -613,13 +613,6 @@ static int azx_pcm_open(struct snd_pcm_s20 , 178000000); -/* by some reason, the playback stream stalls on PulseAudio with -* tsched=1 when a capture stream triggers. Until we figure out the -* real cause, disable tsched mode by telling the PCM info flag. -*/ -if (chip->driver_caps &AZX_DCAPS_AMD_WORKAROUND ) - runtime->hw.info |=SNDRV_PCM_INFO_BATCH ; -if (chip->align_buffer_size)/* constrain buffer sizes to be multiple of 128 bytes. This is more efficient in terms of memory
[email protected]
maintainerPATCH 92/168 ALSA: hda: Flush pending unsolicited events before suspend
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Takashi Iwai <[email protected]>
commit 13661fc48461282e43fe8f76bf5bf449b3d40687 upstream.
The HD-audio controller driver processes the unsolicited events via its work asynchronously, and this might be pending when the system goes to suspend. When a lengthy event handling like ELD byte reads is running, this might trigger unexpected accesses among suspend/resume procedure, typically seen with Nvidia driver that still requires the handling via unsolicited event verbs for ELD updates.
This patch adds the flush of unsol_work to assure that pending events are processed before going into suspend.
--- sound/pci/hda/hda_intel.c | 2 ++ 1 file changed, 2 insertions(+)
sound/pci/hda/hda_intel.c +2 -0
@@ -1025,6 +1025,8 @@ static int azx_prepare(struct device *de chip = card->private_data; chip->pm_prepared =1 ; +flush_work (&azx_bus (chip)->unsol_work); +/* HDA controller always requires different WAKEEN for runtime suspend * and system suspend, so don't use direct-complete here. */
[email protected]
maintainerPATCH 93/168 ALSA: hda: Avoid spurious unsol event handling during S3/S4
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Takashi Iwai <[email protected]>
commit 5ff9dde42e8c72ed8102eb8cb62e03f9dc2103ab upstream.
When HD-audio bus receives unsolicited events during its system suspend/resume (S3 and S4) phase, the controller driver may still try to process events although the codec chips are already (or yet) powered down. This might screw up the codec communication, resulting in CORB/RIRB errors. Such events should be rather skipped, as the codec chip status such as the jack status will be fully refreshed at the system resume time.
Since we're tracking the system suspend/resume state in codec power.power_state field, let's add the check in the common unsol event handler entry point to filter out such events.
--- sound/pci/hda/hda_bind.c | 4 ++++ 1 file changed, 4 insertions(+)
sound/pci/hda/hda_bind.c +4 -0
@@ -47,6 +47,10 @@ static void hda_codec_unsol_event(structif (codec->bus->shutdown)return ; +/* ignore unsol events during system suspend/resume */ +if (codec->core.dev.power.power_state.event !=PM_EVENT_ON ) +return ; +if (codec->patch_ops.unsol_event) codec->patch_ops.unsol_event (codec, ev); }
[email protected]
maintainerPATCH 94/168 ALSA: usb-audio: Fix "cannot get freq eq" errors on Dell AE515 sound bar
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Takashi Iwai <[email protected]>
commit fec60c3bc5d1713db2727cdffc638d48f9c07dc3 upstream.
Dell AE515 sound bar (413c:a506) spews the error messages when the driver tries to read the current sample frequency, hence it needs to be on the list in snd_usb_get_sample_rate_quirk().
--- sound/usb/quirks.c | 1 + 1 file changed, 1 insertion(+)
sound/usb/quirks.c +1 -0
@@ -1452,6 +1452,7 @@ bool snd_usb_get_sample_rate_quirk(struccase USB_ID (0x1901 ,0x0191 ):/* GE B850V3 CP2114 audio interface */ case USB_ID (0x21B4 ,0x0081 ):/* AudioQuest DragonFly */ case USB_ID (0x2912 ,0x30c8 ):/* Audioengine D1 */ +case USB_ID (0x413c ,0xa506 ):/* Dell AE515 sound bar */ return true; }
[email protected]
maintainerPATCH 95/168 ALSA: usb-audio: Apply the control quirk to Plantronics headsets
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Takashi Iwai <[email protected]>
commit 06abcb18b3a021ba1a3f2020cbefb3ed04e59e72 upstream.
Other Plantronics headset models seem requiring the same workaround as C320-M to add the 20ms delay for the control messages, too. Apply the workaround generically for devices with the vendor ID 0x047f.
Note that the problem didn't surface before 5.11 just with luck. Since 5.11 got a big code rewrite about the stream handling, the parameter setup procedure has changed, and this seemed triggering the problem more often.
--- sound/usb/quirks.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
sound/usb/quirks.c +3 -3
@@ -1606,10 +1606,10 @@ void snd_usb_ctl_msg_quirk(struct usb_demsleep (20 );/* -* Plantronics C320-M needs a delay to avoid random -* microhpone failures. +* Plantronics headsets (C320, C320-M, etc) need a delay to avoid +* random microhpone failures. */ -if (chip->usb_id ==USB_ID (0x047f ,0xc025 ) && +if (USB_ID_VENDOR (chip->usb_id) ==0x047f && (requesttype &USB_TYPE_MASK ) ==USB_TYPE_CLASS )msleep (20 );
[email protected]
maintainerPATCH 96/168 Revert 95ebabde382c ("capabilities: Dont allow writing ambiguous v3 file capabilities")
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Eric W. Biederman <[email protected]>
commit 3b0c2d3eaa83da259d7726192cf55a137769012f upstream.
It turns out that there are in fact userspace implementations that care and this recent change caused a regression.
https://github.com/containers/buildah/issues/3071
As the motivation for the original change was future development, and the impact is existing real world code just revert this change and allow the ambiguity in v3 file caps.
--- security/commoncap.c | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-)
security/commoncap.c +1 -11
@@ -500,8 +500,7 @@ int cap_convert_nscap(struct dentry *den__u32 magic, nsmagic;struct inode *inode =d_backing_inode (dentry);struct user_namespace *task_ns =current_user_ns (), - *fs_ns = inode->i_sb->s_user_ns, - *ancestor; + *fs_ns = inode->i_sb->s_user_ns;kuid_t rootid;size_t newsize;
@@ -524,15 +523,6 @@ int cap_convert_nscap(struct dentry *denif (nsrootid ==-1 )return -EINVAL ; -/* -* Do not allow allow adding a v3 filesystem capability xattr -* if the rootid field is ambiguous. -*/ -for (ancestor = task_ns->parent; ancestor; ancestor = ancestor->parent) { -if (from_kuid (ancestor, rootid) ==0 ) -return -EINVAL ; - } - newsize =sizeof (struct vfs_ns_cap_data ); nscap =kmalloc (newsize,GFP_ATOMIC );if (!nscap)
[email protected]
maintainerPATCH 97/168 arm64: kasan: fix page_alloc tagging with DEBUG_VIRTUAL
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Andrey Konovalov <[email protected]>
commit 86c83365ab76e4b43cedd3ce07a07d32a4dc79ba upstream.
When CONFIG_DEBUG_VIRTUAL is enabled, the default page_to_virt() macro implementation from include/linux/mm.h is used. That definition doesn't account for KASAN tags, which leads to no tags on page_alloc allocations.
Provide an arm64-specific definition for page_to_virt() when CONFIG_DEBUG_VIRTUAL is enabled that takes care of KASAN tags.
--- arch/arm64/include/asm/memory.h | 5 +++++ 1 file changed, 5 insertions(+)
arch/arm64/include/asm/memory.h +5 -0
@@ -315,6 +315,11 @@ static inline void *phys_to_virt(phys_ad#define ARCH_PFN_OFFSET ((unsigned long)PHYS_PFN_OFFSET)#if !defined(CONFIG_SPARSEMEM_VMEMMAP ) || defined(CONFIG_DEBUG_VIRTUAL ) +#define page_to_virt (x) ({ \ + __typeof__(x) __page = x; \ + void *__addr = __va(page_to_phys(__page)); \ + (void *)__tag_set((const void *)__addr, page_kasan_tag(__page));\ +})#define virt_to_page (x) pfn_to_page(virt_to_pfn(x))#else #define page_to_virt (x) ({ \
[email protected]
maintainerPATCH 98/168 s390/dasd: fix hanging DASD driver unbind
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Stefan Haberland <[email protected]>
commit 7d365bd0bff3c0310c39ebaffc9a8458e036d666 upstream.
In case of an unbind of the DASD device driver the function dasd_generic_remove() is called which shuts down the device. Among others this functions removes the int_handler from the cdev. During shutdown the device cancels all outstanding IO requests and waits for completion of the clear request.
Unfortunately the clear interrupt will never be received when there is no interrupt handler connected.
Fix by moving the int_handler removal after the call to the state machine where no request or interrupt is outstanding.
--- drivers/s390/block/dasd.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/s390/block/dasd.c +1 -2
@@ -3522,8 +3522,6 @@ void dasd_generic_remove(struct ccw_devistruct dasd_device *device;struct dasd_block *block; - cdev->handler =NULL ; - device =dasd_device_from_cdev (cdev);if (IS_ERR (device)) {dasd_remove_sysfs_files (cdev);
@@ -3542,6 +3540,7 @@ void dasd_generic_remove(struct ccw_devi* no quite down yet. */ dasd_set_target_state (device,DASD_STATE_NEW ); + cdev->handler =NULL ;/* dasd_delete_device destroys the device reference. */ block = device->block;dasd_delete_device (device);
[email protected]
maintainerPATCH 99/168 s390/dasd: fix hanging IO request during DASD driver unbind
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Stefan Haberland <[email protected]>
commit 66f669a272898feb1c69b770e1504aa2ec7723d1 upstream.
Prevent that an IO request is build during device shutdown initiated by a driver unbind. This request will never be able to be processed or canceled and will hang forever. This will lead also to a hanging unbind.
Fix by checking not only if the device is in READY state but also check that there is no device offline initiated before building a new IO request.
--- drivers/s390/block/dasd.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/s390/block/dasd.c +2 -1
@@ -3087,7 +3087,8 @@ static blk_status_t do_dasd_request(stru basedev = block->base;spin_lock_irq (&dq->lock); -if (basedev->state <DASD_STATE_READY ) { +if (basedev->state <DASD_STATE_READY || +test_bit (DASD_FLAG_OFFLINE , &basedev->flags)) {DBF_DEV_EVENT (DBF_ERR , basedev,"device not ready for request %p" , req); rc =BLK_STS_IOERR ;
[email protected]
maintainerPATCH 100/168 software node: Fix node registration
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Heikki Krogerus <[email protected]>
commit 8891123f9cbb9c1ee531e5a87fa116f0af685c48 upstream.
Software node can not be registered before its parent.
--- drivers/base/swnode.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/base/swnode.c +3 -0
@@ -812,6 +812,9 @@ int software_node_register(const structif (software_node_to_swnode (node))return -EEXIST ; +if (node->parent && !parent) +return -EINVAL ; +return PTR_ERR_OR_ZERO (swnode_register (node, parent,0 )); }EXPORT_SYMBOL_GPL (software_node_register);
[email protected]
maintainerPATCH 101/168 mmc: core: Fix partition switch time for eMMC
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Adrian Hunter <[email protected]>
commit 66fbacccbab91e6e55d9c8f1fc0910a8eb6c81f7 upstream.
Avoid the following warning by always defining partition switch time:
[ 3.209874] mmc1: unspecified timeout for CMD6 - use generic [ 3.222780] ------------[ cut here ]------------ [ 3.233363] WARNING: CPU: 1 PID: 111 at drivers/mmc/core/mmc_ops.c:575 __mmc_switch+0x200/0x204
--- drivers/mmc/core/mmc.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-)
drivers/mmc/core/mmc.c +11 -4
@@ -423,10 +423,6 @@ static int mmc_decode_ext_csd(struct mmc/* EXT_CSD value is in units of 10ms, but we store in ms */ card->ext_csd.part_time =10 * ext_csd[EXT_CSD_PART_SWITCH_TIME ]; -/* Some eMMC set the value too low so set a minimum */ -if (card->ext_csd.part_time && - card->ext_csd.part_time <MMC_MIN_PART_SWITCH_TIME ) - card->ext_csd.part_time =MMC_MIN_PART_SWITCH_TIME ;/* Sleep / awake timeout in 100ns units */ if (sa_shift >0 && sa_shift <=0x17 )
@@ -616,6 +612,17 @@ static int mmc_decode_ext_csd(struct mmc card->ext_csd.data_sector_size =512 ; } +/* +* GENERIC_CMD6_TIME is to be used "unless a specific timeout is defined +* when accessing a specific field", so use it here if there is no +* PARTITION_SWITCH_TIME. +*/ +if (!card->ext_csd.part_time) + card->ext_csd.part_time = card->ext_csd.generic_cmd6_time; +/* Some eMMC set the value too low so set a minimum */ +if (card->ext_csd.part_time <MMC_MIN_PART_SWITCH_TIME ) + card->ext_csd.part_time =MMC_MIN_PART_SWITCH_TIME ; +/* eMMC v5 or later */ if (card->ext_csd.rev >=7 ) { memcpy(card->ext_csd.fwrev, &ext_csd[EXT_CSD_FIRMWARE_VERSION],
[email protected]
maintainerPATCH 102/168 mmc: cqhci: Fix random crash when remove mmc module/card
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Frank Li <[email protected]>
commit f06391c45e83f9a731045deb23df7cc3814fd795 upstream.
[ 6684.493350] Unable to handle kernel paging request at virtual address ffff800011c5b0f0 [ 6684.498531] mmc0: card 0001 removed [ 6684.501556] Mem abort info: [ 6684.509681] ESR = 0x96000047 [ 6684.512786] EC = 0x25: DABT (current EL), IL = 32 bits [ 6684.518394] SET = 0, FnV = 0 [ 6684.521707] EA = 0, S1PTW = 0 [ 6684.524998] Data abort info: [ 6684.528236] ISV = 0, ISS = 0x00000047 [ 6684.532986] CM = 0, WnR = 1 [ 6684.536129] swapper pgtable: 4k pages, 48-bit VAs, pgdp=0000000081b22000 [ 6684.543923] [ffff800011c5b0f0] pgd=00000000bffff003, p4d=00000000bffff003, pud=00000000bfffe003, pmd=00000000900e1003, pte=0000000000000000 [ 6684.557915] Internal error: Oops: 96000047 [#1] PREEMPT SMP [ 6684.564240] Modules linked in: sdhci_esdhc_imx(-) sdhci_pltfm sdhci cqhci mmc_block mmc_core fsl_jr_uio caam_jr caamkeyblob_desc caamhash_desc caamalg_desc crypto_engine rng_core authenc libdes crct10dif_ce flexcan can_dev caam error [last unloaded: mmc_core] [ 6684.587281] CPU: 0 PID: 79138 Comm: kworker/0:3H Not tainted 5.10.9-01410-g3ba33182767b-dirty #10 [ 6684.596160] Hardware name: Freescale i.MX8DXL EVK (DT) [ 6684.601320] Workqueue: kblockd blk_mq_run_work_fn
[ 6684.606094] pstate: 40000005 (nZcv daif -PAN -UAO -TCO BTYPE=--) [ 6684.612286] pc : cqhci_request+0x148/0x4e8 [cqhci] ^GMessage from syslogd@ at Thu Jan 1 01:51:24 1970 ...[ 6684.617085] lr : cqhci_request+0x314/0x4e8 [cqhci] [ 6684.626734] sp : ffff80001243b9f0 [ 6684.630049] x29: ffff80001243b9f0 x28: ffff00002c3dd000 [ 6684.635367] x27: 0000000000000001 x26: 0000000000000001 [ 6684.640690] x25: ffff00002c451000 x24: 000000000000000f [ 6684.646007] x23: ffff000017e71c80 x22: ffff00002c451000 [ 6684.651326] x21: ffff00002c0f3550 x20: ffff00002c0f3550 [ 6684.656651] x19: ffff000017d46880 x18: ffff00002cea1500 [ 6684.661977] x17: 0000000000000000 x16: 0000000000000000 [ 6684.667294] x15: 000001ee628e3ed1 x14: 0000000000000278 [ 6684.672610] x13: 0000000000000001 x12: 0000000000000001 [ 6684.677927] x11: 0000000000000000 x10: 0000000000000000 [ 6684.683243] x9 : 000000000000002b x8 : 0000000000001000 [ 6684.688560] x7 : 0000000000000010 x6 : ffff00002c0f3678 [ 6684.693886] x5 : 000000000000000f x4 : ffff800011c5b000 [ 6684.699211] x3 : 000000000002d988 x2 : 0000000000000008 [ 6684.704537] x1 : 00000000000000f0 x0 : 0002d9880008102f [ 6684.709854] Call trace: [ 6684.712313] cqhci_request+0x148/0x4e8 [cqhci] [ 6684.716803] mmc_cqe_start_req+0x58/0x68 [mmc_core] [ 6684.721698] mmc_blk_mq_issue_rq+0x460/0x810 [mmc_block] [ 6684.727018] mmc_mq_queue_rq+0x118/0x2b0 [mmc_block]
The problem occurs when cqhci_request() get called after cqhci_disable() as it leads to access of allocated memory that has already been freed. Let's fix the problem by calling cqhci_disable() a bit later in the remove path.
--- drivers/mmc/core/bus.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/mmc/core/bus.c +5 -6
@@ -373,11 +373,6 @@ void mmc_remove_card(struct mmc_card *cammc_remove_card_debugfs (card);#endif -if (host->cqe_enabled) { - host->cqe_ops->cqe_disable (host); - host->cqe_enabled = false; - } -if (mmc_card_present (card)) {if (mmc_host_is_spi (card->host)) { pr_info("%s: SPI card removed\n",
@@ -390,6 +385,10 @@ void mmc_remove_card(struct mmc_card *caof_node_put (card->dev.of_node); } +if (host->cqe_enabled) { + host->cqe_ops->cqe_disable (host); + host->cqe_enabled = false; + } +put_device (&card->dev); } -
[email protected]
maintainerPATCH 103/168 Goodix Fingerprint device is not a modem
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Yorick de Wid <[email protected]>
commit 4d8654e81db7346f915eca9f1aff18f385cab621 upstream.
The CDC ACM driver is false matching the Goodix Fingerprint device against the USB_CDC_ACM_PROTO_AT_V25TER.
The Goodix Fingerprint device is a biometrics sensor that should be handled in user-space. libfprint has some support for Goodix fingerprint sensors, although not for this particular one. It is possible that the vendor allocates a PID per OEM (Lenovo, Dell etc). If this happens to be the case then more devices from the same vendor could potentially match the ACM modem module table.
--- drivers/usb/class/cdc-acm.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/usb/class/cdc-acm.c +5 -0
@@ -1941,6 +1941,11 @@ static const struct usb_device_id acm_id .driver_info = SEND_ZERO_PACKET, }, +/* Exclude Goodix Fingerprint Reader */ + { USB_DEVICE(0x27c6, 0x5395), + .driver_info =IGNORE_DEVICE , + }, + /* control interfaces without any protocol set */ { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_ACM, USB_CDC_PROTO_NONE) },
[email protected]
maintainerPATCH 104/168 USB: gadget: u_ether: Fix a configfs return code
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Dan Carpenter <[email protected]>
commit 650bf52208d804ad5ee449c58102f8dc43175573 upstream.
If the string is invalid, this should return -EINVAL instead of 0.
--- drivers/usb/gadget/function/u_ether_configfs.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/usb/gadget/function/u_ether_configfs.h +2 -3
@@ -169,12 +169,11 @@ out: \ size_t len) \ { \ struct f_##_f_##_opts *opts = to_f_##_f_##_opts(item); \ -int ret; \ +int ret = -EINVAL ; \u8 val; \ \mutex_lock (&opts->lock); \ - ret =sscanf (page,"%02hhx" , &val); \ -if (ret >0 ) { \ +if (sscanf (page,"%02hhx" , &val) >0 ) { \ opts->_n_ = val; \ ret = len; \ } \
[email protected]
maintainerPATCH 105/168 usb: gadget: f_uac2: always increase endpoint max_packet_size by one audio slot
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ruslan Bilovol <[email protected]>
commit 789ea77310f0200c84002884ffd628e2baf3ad8a upstream.
As per UAC2 Audio Data Formats spec (2.3.1.1 USB Packets), if the sampling rate is a constant, the allowable variation of number of audio slots per virtual frame is +/- 1 audio slot.
It means that endpoint should be able to accept/send +1 audio slot.
Previous endpoint max_packet_size calculation code was adding sometimes +1 audio slot due to DIV_ROUND_UP behaviour which was rounding up to closest integer. However this doesn't work if the numbers are divisible.
It had no any impact with Linux hosts which ignore this issue, but in case of more strict Windows it caused rejected enumeration
Thus always add +1 audio slot to endpoint's max packet size
--- drivers/usb/gadget/function/f_uac2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/gadget/function/f_uac2.c +1 -1
@@ -478,7 +478,7 @@ static int set_ep_max_packet_size(const } max_size_bw =num_channels (chmask) * ssize * -DIV_ROUND_UP (srate, factor / (1 << (ep_desc->bInterval -1 ))); + ((srate / (factor / (1 << (ep_desc->bInterval -1 )))) +1 ); ep_desc->wMaxPacketSize =cpu_to_le16 (min_t (u16, max_size_bw, max_size_ep));
[email protected]
maintainerPATCH 106/168 usb: gadget: f_uac1: stop playback on function disable
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ruslan Bilovol <[email protected]>
commit cc2ac63d4cf72104e0e7f58bb846121f0f51bb19 upstream.
There is missing playback stop/cleanup in case of gadget's ->disable callback that happens on events like USB host resetting or gadget disconnection
--- drivers/usb/gadget/function/f_uac1.c | 1 + 1 file changed, 1 insertion(+)
drivers/usb/gadget/function/f_uac1.c +1 -0
@@ -499,6 +499,7 @@ static void f_audio_disable(struct usb_f uac1->as_out_alt =0 ; uac1->as_in_alt =0 ; +u_audio_stop_playback (&uac1->g_audio);u_audio_stop_capture (&uac1->g_audio); }
[email protected]
maintainerPATCH 107/168 usb: dwc3: qcom: Add missing DWC3 OF node refcount decrement
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Serge Semin <[email protected]>
commit 1cffb1c66499a9db9a735473778abf8427d16287 upstream.
of_get_child_by_name() increments the reference counter of the OF node it managed to find. So after the code is done using the device node, the refcount must be decremented. Add missing of_node_put() invocation then to the dwc3_qcom_of_register_core() method, since DWC3 OF node is being used only there.
--- drivers/usb/dwc3/dwc3-qcom.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/usb/dwc3/dwc3-qcom.c +6 -3
@@ -528,16 +528,19 @@ static int dwc3_qcom_of_register_core(st ret =of_platform_populate (np,NULL ,NULL , dev);if (ret) {dev_err (dev,"failed to register dwc3 core - %d\n" , ret); -return ret; + goto node_put; } qcom->dwc3 =of_find_device_by_node (dwc3_np);if (!qcom->dwc3) { + ret = -ENODEV ;dev_err (dev,"failed to get dwc3 platform device\n" ); -return -ENODEV ; } -return 0 ; +node_put: +of_node_put (dwc3_np); + +return ret; }static const struct dwc3_acpi_pdata sdm845_acpi_pdata = {
[email protected]
maintainerPATCH 108/168 usb: dwc3: qcom: Honor wakeup enabled/disabled state
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Matthias Kaehlcke <[email protected]>
commit 2664deb0930643149d61cddbb66ada527ae180bd upstream.
The dwc3-qcom currently enables wakeup interrupts unconditionally when suspending, however this should not be done when wakeup is disabled (e.g. through the sysfs attribute power/wakeup). Only enable wakeup interrupts when device_may_wakeup() returns true.
--- drivers/usb/dwc3/dwc3-qcom.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/usb/dwc3/dwc3-qcom.c +5 -2
@@ -251,8 +251,10 @@ static int dwc3_qcom_suspend(struct dwc3for (i = qcom->num_clocks -1 ; i >=0 ; i--)clk_disable_unprepare (qcom->clks[i]); +if (device_may_wakeup (qcom->dev)) +dwc3_qcom_enable_interrupts (qcom); + qcom->is_suspended = true; -dwc3_qcom_enable_interrupts (qcom);return 0 ; }
@@ -265,7 +267,8 @@ static int dwc3_qcom_resume(struct dwc3_if (!qcom->is_suspended)return 0 ; -dwc3_qcom_disable_interrupts (qcom); +if (device_may_wakeup (qcom->dev)) +dwc3_qcom_disable_interrupts (qcom);for (i =0 ; i < qcom->num_clocks; i++) { ret =clk_prepare_enable (qcom->clks[i]);
[email protected]
maintainerPATCH 109/168 USB: usblp: fix a hang in poll() if disconnected
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Pete Zaitcev <[email protected]>
commit 9de2c43acf37a17dc4c69ff78bb099b80fb74325 upstream.
Apparently an application that opens a device and calls select() on it, will hang if the decice is disconnected. It's a little surprising that we had this bug for 15 years, but apparently nobody ever uses select() with a printer: only write() and read(), and those work fine. Well, you can also select() with a timeout.
The fix is modeled after devio.c. A few other drivers check the condition first, then do not add the wait queue in case the device is disconnected. We doubt that's completely race-free. So, this patch adds the process first, then locks properly and checks for the disconnect.
--- drivers/usb/class/usblp.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)
drivers/usb/class/usblp.c +12 -4
@@ -494,16 +494,24 @@ static int usblp_release(struct inode *i/* No kernel lock - fine */ static __poll_t usblp_poll (struct file *file,struct poll_table_struct *wait) { -__poll_t ret; +struct usblp *usblp = file->private_data; +__poll_t ret =0 ;unsigned long flags; -struct usblp *usblp = file->private_data;/* Should we check file->f_mode & FMODE_WRITE before poll_wait()? */ poll_wait (file, &usblp->rwait, wait);poll_wait (file, &usblp->wwait, wait); + +mutex_lock (&usblp->mut); +if (!usblp->present) + ret |=EPOLLHUP ; +mutex_unlock (&usblp->mut); +spin_lock_irqsave (&usblp->lock, flags); - ret = ((usblp->bidir && usblp->rcomplete) ?EPOLLIN |EPOLLRDNORM :0 ) | - ((usblp->no_paper || usblp->wcomplete) ?EPOLLOUT |EPOLLWRNORM :0 ); +if (usblp->bidir && usblp->rcomplete) + ret |=EPOLLIN |EPOLLRDNORM ; +if (usblp->no_paper || usblp->wcomplete) + ret |=EPOLLOUT |EPOLLWRNORM ;spin_unlock_irqrestore (&usblp->lock, flags);return ret; }
[email protected]
maintainerPATCH 110/168 usb: renesas_usbhs: Clear PIPECFG for re-enabling pipe with other EPNUM
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Yoshihiro Shimoda <[email protected]>
commit b1d25e6ee57c2605845595b6c61340d734253eb3 upstream.
According to the datasheet, this controller has a restriction which "set an endpoint number so that combinations of the DIR bit and the EPNUM bits do not overlap.". However, since the udc core driver is possible to assign a bulk pipe as an interrupt endpoint, an endpoint number may not match the pipe number. After that, when user rebinds another gadget driver, this driver broke the restriction because the driver didn't clear any configuration in usb_ep_disable().
Example:
# modprobe g_ncm Then, EP3 = pipe 3, EP4 = pipe 4, EP5 = pipe 6 # rmmod g_ncm # modprobe g_hid Then, EP3 = pipe 6, EP4 = pipe 7. So, pipe 3 and pipe 6 are set as EP3.
So, clear PIPECFG register in usbhs_pipe_free().
--- drivers/usb/renesas_usbhs/pipe.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/usb/renesas_usbhs/pipe.c +2 -0
@@ -746,6 +746,8 @@ struct usbhs_pipe *usbhs_pipe_malloc(strvoid usbhs_pipe_free (struct usbhs_pipe *pipe) { +usbhsp_pipe_select (pipe); +usbhsp_pipe_cfg_set (pipe,0xFFFF ,0 );usbhsp_put_pipe (pipe); }
[email protected]
maintainerPATCH 111/168 usb: xhci: do not perform Soft Retry for some xHCI hosts
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Stanislaw Gruszka <[email protected]>
commit a4a251f8c23518899d2078c320cf9ce2fa459c9f upstream.
On some systems rt2800usb and mt7601u devices are unable to operate since commit f8f80be501aa ("xhci: Use soft retry to recover faster from transaction errors")
Seems that some xHCI controllers can not perform Soft Retry correctly, affecting those devices.
To avoid the problem add xhci->quirks flag that restore pre soft retry xhci behaviour for affected xHCI controllers. Currently those are AMD_PROMONTORYA_4 and AMD_PROMONTORYA_2, since it was confirmed by the users: on those xHCI hosts issue happen and is gone after disabling Soft Retry.
[minor commit message rewording for checkpatch -Mathias]
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=202541
--- drivers/usb/host/xhci-pci.c | 5 +++++ drivers/usb/host/xhci-ring.c | 3 ++- drivers/usb/host/xhci.h | 1 + 3 files changed, 8 insertions(+), 1 deletion(-)
drivers/usb/host/xhci-pci.c +5 -0
@@ -277,6 +277,11 @@ static void xhci_pci_quirks(struct devic pdev->device == 0x9026) xhci->quirks |=XHCI_RESET_PLL_ON_DISCONNECT ; +if (pdev->vendor ==PCI_VENDOR_ID_AMD && + (pdev->device ==PCI_DEVICE_ID_AMD_PROMONTORYA_2 || + pdev->device ==PCI_DEVICE_ID_AMD_PROMONTORYA_4 )) + xhci->quirks |=XHCI_NO_SOFT_RETRY ; +if (xhci->quirks &XHCI_RESET_ON_RESUME )xhci_dbg_trace (xhci, trace_xhci_dbg_quirks,"QUIRK: Resetting on resume" );
drivers/usb/host/xhci-ring.c +2 -1
@@ -2299,7 +2299,8 @@ static int process_bulk_intr_td(struct x remaining =0 ;break ;case COMP_USB_TRANSACTION_ERROR : -if ((ep_ring->err_count++ >MAX_SOFT_RETRY ) || +if (xhci->quirks &XHCI_NO_SOFT_RETRY || + (ep_ring->err_count++ >MAX_SOFT_RETRY ) ||le32_to_cpu (slot_ctx->tt_info) &TT_SLOT )break ; *status =0 ;
drivers/usb/host/xhci.h +1 -0
@@ -1875,6 +1875,7 @@ struct xhci_hcd {#define XHCI_SNPS_BROKEN_SUSPEND BIT_ULL(35)#define XHCI_SKIP_PHY_INIT BIT_ULL(37)#define XHCI_DISABLE_SPARSE BIT_ULL(38) +#define XHCI_NO_SOFT_RETRY BIT_ULL(40)unsigned int num_active_eps;unsigned int limit_active_eps;
[email protected]
maintainerPATCH 112/168 xhci: Improve detection of device initiated wake signal.
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Mathias Nyman <[email protected]>
commit 253f588c70f66184b1f3a9bbb428b49bbda73e80 upstream.
A xHC USB 3 port might miss the first wake signal from a USB 3 device if the port LFPS reveiver isn't enabled fast enough after xHC resume.
xHC host will anyway be resumed by a PME# signal, but will go back to suspend if no port activity is seen.
The device resends the U3 LFPS wake signal after a 100ms delay, but by then host is already suspended, starting all over from the beginning of this issue.
USB 3 specs say U3 wake LFPS signal is sent for max 10ms, then device needs to delay 100ms before resending the wake.
Don't suspend immediately if port activity isn't detected in resume. Instead add a retry. If there is no port activity then delay for 120ms, and re-check for port activity.
--- drivers/usb/host/xhci.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/usb/host/xhci.c +13 -3
@@ -1088,6 +1088,7 @@ int xhci_resume(struct xhci_hcd *xhci, bstruct usb_hcd *secondary_hcd;int retval =0 ;bool comp_timer_running = false; +bool pending_portevent = false;if (!hcd->state)return 0 ;
@@ -1226,13 +1227,22 @@ int xhci_resume(struct xhci_hcd *xhci, b done:if (retval ==0 ) { -/* Resume root hubs only when have pending events. */ -if (xhci_pending_portevent (xhci)) { +/* +* Resume roothubs only if there are pending events. +* USB 3 devices resend U3 LFPS wake after a 100ms delay if +* the first wake signalling failed, give it that chance. +*/ + pending_portevent =xhci_pending_portevent (xhci); +if (!pending_portevent) { +msleep (120 ); + pending_portevent =xhci_pending_portevent (xhci); + } + +if (pending_portevent) {usb_hcd_resume_root_hub (xhci->shared_hcd);usb_hcd_resume_root_hub (hcd); } } -/* * If system is subject to the Quirk, Compliance Mode Timer needs to * be re-initialized Always after a system resume. Ports are subject
[email protected]
maintainerPATCH 113/168 usb: xhci: Fix ASMedia ASM1042A and ASM3242 DMA addressing
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Forest Crossman <[email protected]>
commit b71c669ad8390dd1c866298319ff89fe68b45653 upstream.
I've confirmed that both the ASMedia ASM1042A and ASM3242 have the same problem as the ASM1142 and ASM2142/ASM3142, where they lose some of the upper bits of 64-bit DMA addresses. As with the other chips, this can cause problems on systems where the upper bits matter, and adding the XHCI_NO_64BIT_SUPPORT quirk completely fixes the issue.
--- drivers/usb/host/xhci-pci.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/usb/host/xhci-pci.c +6 -2
@@ -62,6 +62,7 @@#define PCI_DEVICE_ID_ASMEDIA_1042A_XHCI 0x1142#define PCI_DEVICE_ID_ASMEDIA_1142_XHCI 0x1242#define PCI_DEVICE_ID_ASMEDIA_2142_XHCI 0x2142 +#define PCI_DEVICE_ID_ASMEDIA_3242_XHCI 0x3242static const char hcd_name[] ="xhci_hcd" ;
@@ -258,11 +259,14 @@ static void xhci_pci_quirks(struct devic pdev->device == PCI_DEVICE_ID_ASMEDIA_1042_XHCI) xhci->quirks |= XHCI_BROKEN_STREAMS;if (pdev->vendor ==PCI_VENDOR_ID_ASMEDIA && - pdev->device ==PCI_DEVICE_ID_ASMEDIA_1042A_XHCI ) + pdev->device ==PCI_DEVICE_ID_ASMEDIA_1042A_XHCI ) { xhci->quirks |=XHCI_TRUST_TX_LENGTH ; + xhci->quirks |=XHCI_NO_64BIT_SUPPORT ; + }if (pdev->vendor ==PCI_VENDOR_ID_ASMEDIA && (pdev->device ==PCI_DEVICE_ID_ASMEDIA_1142_XHCI || - pdev->device ==PCI_DEVICE_ID_ASMEDIA_2142_XHCI )) + pdev->device ==PCI_DEVICE_ID_ASMEDIA_2142_XHCI || + pdev->device ==PCI_DEVICE_ID_ASMEDIA_3242_XHCI )) xhci->quirks |=XHCI_NO_64BIT_SUPPORT ; if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&
[email protected]
maintainerPATCH 114/168 xhci: Fix repeated xhci wake after suspend due to uncleared internal wake state
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Mathias Nyman <[email protected]>
commit d26c00e7276fc92b18c253d69e872f6b03832bad upstream.
If port terminations are detected in suspend, but link never reaches U0 then xHCI may have an internal uncleared wake state that will cause an immediate wake after suspend.
This wake state is normally cleared when driver clears the PORT_CSC bit, which is set after a device is enabled and in U0.
Write 1 to clear PORT_CSC for ports that don't have anything connected when suspending. This makes sure any pending internal wake states in xHCI are cleared.
--- drivers/usb/host/xhci.c | 62 +++++++++++++++++++++++------------------------- 1 file changed, 30 insertions(+), 32 deletions(-)
drivers/usb/host/xhci.c +30 -32
@@ -883,44 +883,42 @@ static void xhci_clear_command_ring(struxhci_set_cmd_ring_deq (xhci); } -static void xhci_disable_port_wake_on_bits (struct xhci_hcd *xhci) +/* +* Disable port wake bits if do_wakeup is not set. +* +* Also clear a possible internal port wake state left hanging for ports that +* detected termination but never successfully enumerated (trained to 0U). +* Internal wake causes immediate xHCI wake after suspend. PORT_CSC write done +* at enumeration clears this wake, force one here as well for unconnected ports +*/ + +static void xhci_disable_hub_port_wake (struct xhci_hcd *xhci, +struct xhci_hub *rhub, +bool do_wakeup) { -struct xhci_port **ports; -int port_index;unsigned long flags;u32 t1, t2, portsc; +int i;spin_lock_irqsave (&xhci->lock, flags); -/* disable usb3 ports Wake bits */ - port_index = xhci->usb3_rhub.num_ports; - ports = xhci->usb3_rhub.ports; -while (port_index--) { - t1 =readl (ports[port_index]->addr); - portsc = t1; - t1 =xhci_port_state_to_neutral (t1); - t2 = t1 & ~PORT_WAKE_BITS ; -if (t1 != t2) { -writel (t2, ports[port_index]->addr); -xhci_dbg (xhci,"disable wake bits port %d-%d, portsc: 0x%x, write: 0x%x\n" , - xhci->usb3_rhub.hcd->self.busnum, - port_index +1 , portsc, t2); - } - } +for (i =0 ; i < rhub->num_ports; i++) { + portsc =readl (rhub->ports[i]->addr); + t1 =xhci_port_state_to_neutral (portsc); + t2 = t1; + +/* clear wake bits if do_wake is not set */ +if (!do_wakeup) + t2 &= ~PORT_WAKE_BITS ; + +/* Don't touch csc bit if connected or connect change is set */ +if (!(portsc & (PORT_CSC |PORT_CONNECT ))) + t2 |=PORT_CSC ; -/* disable usb2 ports Wake bits */ - port_index = xhci->usb2_rhub.num_ports; - ports = xhci->usb2_rhub.ports; -while (port_index--) { - t1 =readl (ports[port_index]->addr); - portsc = t1; - t1 =xhci_port_state_to_neutral (t1); - t2 = t1 & ~PORT_WAKE_BITS ;if (t1 != t2) { -writel (t2, ports[port_index]->addr); -xhci_dbg (xhci,"disable wake bits port %d-%d, portsc: 0x%x, write: 0x%x\n" , - xhci->usb2_rhub.hcd->self.busnum, - port_index +1 , portsc, t2); +writel (t2, rhub->ports[i]->addr); +xhci_dbg (xhci,"config port %d-%d wake bits, portsc: 0x%x, write: 0x%x\n" , + rhub->hcd->self.busnum, i +1 , portsc, t2); } }spin_unlock_irqrestore (&xhci->lock, flags);
@@ -983,8 +981,8 @@ int xhci_suspend(struct xhci_hcd *xhci,return -EINVAL ;/* Clear root port wake on bits if wakeup not allowed. */ -if (!do_wakeup) -xhci_disable_port_wake_on_bits (xhci); +xhci_disable_hub_port_wake (xhci, &xhci->usb3_rhub, do_wakeup); +xhci_disable_hub_port_wake (xhci, &xhci->usb2_rhub, do_wakeup);if (!HCD_HW_ACCESSIBLE (hcd))return 0 ;
[email protected]
maintainerPATCH 115/168 USB: serial: io_edgeport: fix memory leak in edge_startup
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Pavel Skripkin <[email protected]>
commit cfdc67acc785e01a8719eeb7012709d245564701 upstream.
sysbot found memory leak in edge_startup().
The problem was that when an error was received from the usb_submit_urb(), nothing was cleaned up.
--- drivers/usb/serial/io_edgeport.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-)
drivers/usb/serial/io_edgeport.c +16 -10
@@ -3003,26 +3003,32 @@ static int edge_startup(struct usb_seria response = -ENODEV ; } -usb_free_urb (edge_serial->interrupt_read_urb); -kfree (edge_serial->interrupt_in_buffer); - -usb_free_urb (edge_serial->read_urb); -kfree (edge_serial->bulk_in_buffer); - -kfree (edge_serial); - -return response; + goto error; }/* start interrupt read for this edgeport this interrupt will * continue as long as the edgeport is connected */ response =usb_submit_urb (edge_serial->interrupt_read_urb,GFP_KERNEL ); -if (response) +if (response) {dev_err (ddev,"%s - Error %d submitting control urb\n" , __func__, response); + + goto error; + } }return response; + +error: +usb_free_urb (edge_serial->interrupt_read_urb); +kfree (edge_serial->interrupt_in_buffer); + +usb_free_urb (edge_serial->read_urb); +kfree (edge_serial->bulk_in_buffer); + +kfree (edge_serial); + +return response; }
[email protected]
maintainerPATCH 116/168 USB: serial: ch341: add new Product ID
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Niv Sardi <[email protected]>
commit 5563b3b6420362c8a1f468ca04afe6d5f0a8d0a3 upstream.
Add PID for CH340 that's found on cheap programmers.
The driver works flawlessly as soon as the new PID (0x9986) is added to it. These look like ANU232MI but ship with a ch341 inside. They have no special identifiers (mine only has the string "DB9D20130716" printed on the PCB and nothing identifiable on the packaging. The merchant i bought it from doesn't sell these anymore).
the lsusb -v output is:
Bus 001 Device 009: ID 9986:7523
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 1.10
bDeviceClass 255 Vendor Specific Class
bDeviceSubClass 0 bDeviceProtocol 0 bMaxPacketSize0 8 idVendor 0x9986 idProduct 0x7523 bcdDevice 2.54 iManufacturer 0 iProduct 0 iSerial 0 bNumConfigurations 1
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 0x0027
bNumInterfaces 1
bConfigurationValue 1
iConfiguration 0
bmAttributes 0x80
(Bus Powered)
MaxPower 96mA
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 0
bAlternateSetting 0
bNumEndpoints 3
bInterfaceClass 255 Vendor Specific Class
bInterfaceSubClass 1
bInterfaceProtocol 2
iInterface 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x82 EP 2 IN
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0020 1x 32 bytes
bInterval 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x02 EP 2 OUT
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0020 1x 32 bytes
bInterval 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x81 EP 1 IN
bmAttributes 3
Transfer Type Interrupt
Synch Type None
Usage Type Data
wMaxPacketSize 0x0008 1x 8 bytes
bInterval 1
--- drivers/usb/serial/ch341.c | 1 + 1 file changed, 1 insertion(+)
drivers/usb/serial/ch341.c +1 -0
@@ -85,6 +85,7 @@ static const struct usb_device_id id_tab { USB_DEVICE(0x1a86, 0x7522) }, { USB_DEVICE(0x1a86, 0x7523) }, { USB_DEVICE(0x4348, 0x5523) }, + { USB_DEVICE(0x9986, 0x7523) }, { }, };MODULE_DEVICE_TABLE (usb, id_table);
[email protected]
maintainerPATCH 117/168 USB: serial: cp210x: add ID for Acuity Brands nLight Air Adapter
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Karan Singhal <[email protected]>
commit ca667a33207daeaf9c62b106815728718def60ec upstream.
IDs of nLight Air Adapter, Acuity Brands, Inc.:
vid: 10c4
pid: 88d8
--- drivers/usb/serial/cp210x.c | 1 + 1 file changed, 1 insertion(+)
drivers/usb/serial/cp210x.c +1 -0
@@ -146,6 +146,7 @@ static const struct usb_device_id id_tab { USB_DEVICE(0x10C4, 0x8857) }, /* CEL EM357 ZigBee USB Stick */ { USB_DEVICE(0x10C4, 0x88A4) }, /* MMB Networks ZigBee USB Device */ { USB_DEVICE(0x10C4, 0x88A5) }, /* Planet Innovation Ingeni ZigBee USB Device */ + { USB_DEVICE(0x10C4, 0x88D8) }, /* Acuity Brands nLight Air Adapter */ { USB_DEVICE(0x10C4, 0x88FB) }, /* CESINEL MEDCAL STII Network Analyzer */ { USB_DEVICE(0x10C4, 0x8938) }, /* CESINEL MEDCAL S II Network Analyzer */ { USB_DEVICE(0x10C4, 0x8946) }, /* Ketra N1 Wireless Interface */
[email protected]
maintainerPATCH 118/168 USB: serial: cp210x: add some more GE USB IDs
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Sebastian Reichel <[email protected]>
commit 42213a0190b535093a604945db05a4225bf43885 upstream.
GE CS1000 has some more custom USB IDs for CP2102N; add them to the driver to have working auto-probing.
--- drivers/usb/serial/cp210x.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/usb/serial/cp210x.c +2 -0
@@ -203,6 +203,8 @@ static const struct usb_device_id id_tab { USB_DEVICE(0x1901, 0x0194) }, /* GE Healthcare Remote Alarm Box */ { USB_DEVICE(0x1901, 0x0195) }, /* GE B850/B650/B450 CP2104 DP UART interface */ { USB_DEVICE(0x1901, 0x0196) }, /* GE B850 CP2105 DP UART interface */ + { USB_DEVICE(0x1901, 0x0197) }, /* GE CS1000 Display serial interface */ + { USB_DEVICE(0x1901, 0x0198) }, /* GE CS1000 M.2 Key E serial interface */ { USB_DEVICE(0x199B, 0xBA30) }, /* LORD WSDA-200-USB */ { USB_DEVICE(0x19CF, 0x3000) }, /* Parrot NMEA GPS Flight Recorder */ { USB_DEVICE(0x1ADB, 0x0001) }, /* Schweitzer Engineering C662 Cable */
[email protected]
maintainerPATCH 119/168 usbip: fix stub_dev to check for stream socket
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Shuah Khan <[email protected]>
commit 47ccc8fc2c9c94558b27b6f9e2582df32d29e6e8 upstream.
Fix usbip_sockfd_store() to validate the passed in file descriptor is a stream socket. If the file descriptor passed was a SOCK_DGRAM socket, sock_recvmsg() can't detect end of stream.
--- drivers/usb/usbip/stub_dev.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/usb/usbip/stub_dev.c +11 -1
@@ -69,8 +69,16 @@ static ssize_t usbip_sockfd_store(struct } socket =sockfd_lookup (sockfd, &err); -if (!socket) +if (!socket) { +dev_err (dev,"failed to lookup sock" ); goto err; + } + +if (socket->type !=SOCK_STREAM ) { +dev_err (dev,"Expecting SOCK_STREAM - found %d" , + socket->type); + goto sock_err; + } sdev->ud.tcp_socket = socket; sdev->ud.sockfd = sockfd;
@@ -100,6 +108,8 @@ static ssize_t usbip_sockfd_store(structreturn count; +sock_err: +sockfd_put (socket); err:spin_unlock_irq (&sdev->ud.lock);return -EINVAL ;
[email protected]
maintainerPATCH 120/168 usbip: fix vhci_hcd to check for stream socket
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Shuah Khan <[email protected]>
commit f55a0571690c4aae03180e001522538c0927432f upstream.
Fix attach_store() to validate the passed in file descriptor is a stream socket. If the file descriptor passed was a SOCK_DGRAM socket, sock_recvmsg() can't detect end of stream.
--- drivers/usb/usbip/vhci_sysfs.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/usb/usbip/vhci_sysfs.c +9 -1
@@ -349,8 +349,16 @@ static ssize_t attach_store(struct devic/* Extract socket from fd. */ socket =sockfd_lookup (sockfd, &err); -if (!socket) +if (!socket) { +dev_err (dev,"failed to lookup sock" );return -EINVAL ; + } +if (socket->type !=SOCK_STREAM ) { +dev_err (dev,"Expecting SOCK_STREAM - found %d" , + socket->type); +sockfd_put (socket); +return -EINVAL ; + }/* now need lock until setting vdev status as used */
[email protected]
maintainerPATCH 121/168 usbip: fix vudc to check for stream socket
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Shuah Khan <[email protected]>
commit 6801854be94fe8819b3894979875ea31482f5658 upstream.
Fix usbip_sockfd_store() to validate the passed in file descriptor is a stream socket. If the file descriptor passed was a SOCK_DGRAM socket, sock_recvmsg() can't detect end of stream.
--- drivers/usb/usbip/vudc_sysfs.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/usb/usbip/vudc_sysfs.c +9 -0
@@ -138,6 +138,13 @@ static ssize_t usbip_sockfd_store(struct goto unlock_ud; } +if (socket->type !=SOCK_STREAM ) { +dev_err (dev,"Expecting SOCK_STREAM - found %d" , + socket->type); + ret = -EINVAL ; + goto sock_err; + } + udc->ud.tcp_socket = socket;spin_unlock_irq (&udc->ud.lock);
@@ -177,6 +184,8 @@ static ssize_t usbip_sockfd_store(structreturn count; +sock_err: +sockfd_put (socket); unlock_ud:spin_unlock_irq (&udc->ud.lock); unlock:
[email protected]
maintainerPATCH 122/168 usbip: fix stub_dev usbip_sockfd_store() races leading to gpf
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Shuah Khan <[email protected]>
commit 9380afd6df70e24eacbdbde33afc6a3950965d22 upstream.
usbip_sockfd_store() is invoked when user requests attach (import) detach (unimport) usb device from usbip host. vhci_hcd sends import request and usbip_sockfd_store() exports the device if it is free for export.
Export and unexport are governed by local state and shared state
- Shared state (usbip device status, sockfd) - sockfd and Device status are used to determine if stub should be brought up or shut down.
- Local state (tcp_socket, rx and tx thread task_struct ptrs) A valid tcp_socket controls rx and tx thread operations while the device is in exported state.
- While the device is exported, device status is marked used and socket, sockfd, and thread pointers are valid.
Export sequence (stub-up) includes validating the socket and creating receive (rx) and transmit (tx) threads to talk to the client to provide access to the exported device. rx and tx threads depends on local and shared state to be correct and in sync.
Unexport (stub-down) sequence shuts the socket down and stops the rx and tx threads. Stub-down sequence relies on local and shared states to be in sync.
There are races in updating the local and shared status in the current stub-up sequence resulting in crashes. These stem from starting rx and tx threads before local and global state is updated correctly to be in sync.
1. Doesn't handle kthread_create() error and saves invalid ptr in local state that drives rx and tx threads.
2. Updates tcp_socket and sockfd, starts stub_rx and stub_tx threads before updating usbip_device status to SDEV_ST_USED. This opens up a race condition between the threads and usbip_sockfd_store() stub up and down handling.
Fix the above problems:
- Stop using kthread_get_run() macro to create/start threads.
- Create threads and get task struct reference.
- Add kthread_create() failure handling and bail out.
- Hold usbip_device lock to update local and shared states after creating rx and tx threads.
- Update usbip_device status to SDEV_ST_USED.
- Update usbip_device tcp_socket, sockfd, tcp_rx, and tcp_tx
- Start threads after usbip_device (tcp_socket, sockfd, tcp_rx, tcp_tx, and status) is complete.
Credit goes to syzbot and Tetsuo Handa for finding and root-causing the kthread_get_run() improper error handling problem and others. This is a hard problem to find and debug since the races aren't seen in a normal case. Fuzzing forces the race window to be small enough for the kthread_get_run() error path bug and starting threads before updating the local and shared state bug in the stub-up sequence.
Tested with syzbot reproducer:
- https://syzkaller.appspot.com/text?tag=ReproC&x=14801034d00000
--- drivers/usb/usbip/stub_dev.c | 32 +++++++++++++++++++++++++------- 1 file changed, 25 insertions(+), 7 deletions(-)
drivers/usb/usbip/stub_dev.c +25 -7
@@ -46,6 +46,8 @@ static ssize_t usbip_sockfd_store(structint sockfd =0 ;struct socket *socket;int rv; +struct task_struct *tcp_rx =NULL ; +struct task_struct *tcp_tx =NULL ;if (!sdev) {dev_err (dev,"sdev is null\n" );
@@ -80,20 +82,36 @@ static ssize_t usbip_sockfd_store(struct goto sock_err; } - sdev->ud.tcp_socket = socket; - sdev->ud.sockfd = sockfd; - +/* unlock and create threads and get tasks */ spin_unlock_irq (&sdev->ud.lock); + tcp_rx =kthread_create (stub_rx_loop, &sdev->ud,"stub_rx" ); +if (IS_ERR (tcp_rx)) { +sockfd_put (socket); +return -EINVAL ; + } + tcp_tx =kthread_create (stub_tx_loop, &sdev->ud,"stub_tx" ); +if (IS_ERR (tcp_tx)) { +kthread_stop (tcp_rx); +sockfd_put (socket); +return -EINVAL ; + } - sdev->ud.tcp_rx =kthread_get_run (stub_rx_loop, &sdev->ud, -"stub_rx" ); - sdev->ud.tcp_tx =kthread_get_run (stub_tx_loop, &sdev->ud, -"stub_tx" ); +/* get task structs now */ +get_task_struct (tcp_rx); +get_task_struct (tcp_tx); +/* lock and update sdev->ud state */ spin_lock_irq (&sdev->ud.lock); + sdev->ud.tcp_socket = socket; + sdev->ud.sockfd = sockfd; + sdev->ud.tcp_rx = tcp_rx; + sdev->ud.tcp_tx = tcp_tx; sdev->ud.status =SDEV_ST_USED ;spin_unlock_irq (&sdev->ud.lock); +wake_up_process (sdev->ud.tcp_rx); +wake_up_process (sdev->ud.tcp_tx); + }else {dev_info (dev,"stub down\n" );
[email protected]
maintainerPATCH 123/168 usbip: fix vhci_hcd attach_store() races leading to gpf
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Shuah Khan <[email protected]>
commit 718ad9693e3656120064b715fe931f43a6201e67 upstream.
attach_store() is invoked when user requests import (attach) a device from usbip host.
Attach and detach are governed by local state and shared state
- Shared state (usbip device status) - Device status is used to manage the attach and detach operations on import-able devices.
- Local state (tcp_socket, rx and tx thread task_struct ptrs) A valid tcp_socket controls rx and tx thread operations while the device is in exported state.
- Device has to be in the right state to be attached and detached.
Attach sequence includes validating the socket and creating receive (rx) and transmit (tx) threads to talk to the host to get access to the imported device. rx and tx threads depends on local and shared state to be correct and in sync.
Detach sequence shuts the socket down and stops the rx and tx threads. Detach sequence relies on local and shared states to be in sync.
There are races in updating the local and shared status in the current attach sequence resulting in crashes. These stem from starting rx and tx threads before local and global state is updated correctly to be in sync.
1. Doesn't handle kthread_create() error and saves invalid ptr in local state that drives rx and tx threads.
2. Updates tcp_socket and sockfd, starts stub_rx and stub_tx threads before updating usbip_device status to VDEV_ST_NOTASSIGNED. This opens up a race condition between the threads, port connect, and detach handling.
Fix the above problems:
- Stop using kthread_get_run() macro to create/start threads.
- Create threads and get task struct reference.
- Add kthread_create() failure handling and bail out.
- Hold vhci and usbip_device locks to update local and shared states after creating rx and tx threads.
- Update usbip_device status to VDEV_ST_NOTASSIGNED.
- Update usbip_device tcp_socket, sockfd, tcp_rx, and tcp_tx
- Start threads after usbip_device (tcp_socket, sockfd, tcp_rx, tcp_tx, and status) is complete.
Credit goes to syzbot and Tetsuo Handa for finding and root-causing the kthread_get_run() improper error handling problem and others. This is hard problem to find and debug since the races aren't seen in a normal case. Fuzzing forces the race window to be small enough for the kthread_get_run() error path bug and starting threads before updating the local and shared state bug in the attach sequence.
- Update usbip_device tcp_rx and tcp_tx pointers holding vhci and usbip_device locks.
Tested with syzbot reproducer:
- https://syzkaller.appspot.com/text?tag=ReproC&x=14801034d00000
--- drivers/usb/usbip/vhci_sysfs.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-)
drivers/usb/usbip/vhci_sysfs.c +25 -4
@@ -312,6 +312,8 @@ static ssize_t attach_store(struct devicstruct vhci *vhci;int err;unsigned long flags; +struct task_struct *tcp_rx =NULL ; +struct task_struct *tcp_tx =NULL ;/* * @rhport: port number of vhci_hcd
@@ -360,9 +362,24 @@ static ssize_t attach_store(struct devicreturn -EINVAL ; } -/* now need lock until setting vdev status as used */ +/* create threads before locking */ + tcp_rx =kthread_create (vhci_rx_loop, &vdev->ud,"vhci_rx" ); +if (IS_ERR (tcp_rx)) { +sockfd_put (socket); +return -EINVAL ; + } + tcp_tx =kthread_create (vhci_tx_loop, &vdev->ud,"vhci_tx" ); +if (IS_ERR (tcp_tx)) { +kthread_stop (tcp_rx); +sockfd_put (socket); +return -EINVAL ; + } + +/* get task structs now */ +get_task_struct (tcp_rx); +get_task_struct (tcp_tx); -/* begin a lock */ +/* now begin lock until setting vdev status set */ spin_lock_irqsave (&vhci->lock, flags);spin_lock (&vdev->ud.lock);
@@ -372,6 +389,8 @@ static ssize_t attach_store(struct devicspin_unlock_irqrestore (&vhci->lock, flags);sockfd_put (socket); +kthread_stop_put (tcp_rx); +kthread_stop_put (tcp_tx);dev_err (dev,"port %d already used\n" , rhport);/*
@@ -390,14 +409,16 @@ static ssize_t attach_store(struct devic vdev->speed = speed; vdev->ud.sockfd = sockfd; vdev->ud.tcp_socket = socket; + vdev->ud.tcp_rx = tcp_rx; + vdev->ud.tcp_tx = tcp_tx; vdev->ud.status =VDEV_ST_NOTASSIGNED ;spin_unlock (&vdev->ud.lock);spin_unlock_irqrestore (&vhci->lock, flags);/* end the lock */ - vdev->ud.tcp_rx =kthread_get_run (vhci_rx_loop, &vdev->ud,"vhci_rx" ); - vdev->ud.tcp_tx =kthread_get_run (vhci_tx_loop, &vdev->ud,"vhci_tx" ); +wake_up_process (vdev->ud.tcp_rx); +wake_up_process (vdev->ud.tcp_tx);rh_port_connect (vdev, speed);
[email protected]
maintainerPATCH 124/168 usbip: fix vudc usbip_sockfd_store races leading to gpf
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Shuah Khan <[email protected]>
commit 46613c9dfa964c0c60b5385dbdf5aaa18be52a9c upstream.
usbip_sockfd_store() is invoked when user requests attach (import) detach (unimport) usb gadget device from usbip host. vhci_hcd sends import request and usbip_sockfd_store() exports the device if it is free for export.
Export and unexport are governed by local state and shared state
- Shared state (usbip device status, sockfd) - sockfd and Device status are used to determine if stub should be brought up or shut down. Device status is shared between host and client.
- Local state (tcp_socket, rx and tx thread task_struct ptrs) A valid tcp_socket controls rx and tx thread operations while the device is in exported state.
- While the device is exported, device status is marked used and socket, sockfd, and thread pointers are valid.
Export sequence (stub-up) includes validating the socket and creating receive (rx) and transmit (tx) threads to talk to the client to provide access to the exported device. rx and tx threads depends on local and shared state to be correct and in sync.
Unexport (stub-down) sequence shuts the socket down and stops the rx and tx threads. Stub-down sequence relies on local and shared states to be in sync.
There are races in updating the local and shared status in the current stub-up sequence resulting in crashes. These stem from starting rx and tx threads before local and global state is updated correctly to be in sync.
1. Doesn't handle kthread_create() error and saves invalid ptr in local state that drives rx and tx threads.
2. Updates tcp_socket and sockfd, starts stub_rx and stub_tx threads before updating usbip_device status to SDEV_ST_USED. This opens up a race condition between the threads and usbip_sockfd_store() stub up and down handling.
Fix the above problems:
- Stop using kthread_get_run() macro to create/start threads.
- Create threads and get task struct reference.
- Add kthread_create() failure handling and bail out.
- Hold usbip_device lock to update local and shared states after creating rx and tx threads.
- Update usbip_device status to SDEV_ST_USED.
- Update usbip_device tcp_socket, sockfd, tcp_rx, and tcp_tx
- Start threads after usbip_device (tcp_socket, sockfd, tcp_rx, tcp_tx, and status) is complete.
Credit goes to syzbot and Tetsuo Handa for finding and root-causing the kthread_get_run() improper error handling problem and others. This is a hard problem to find and debug since the races aren't seen in a normal case. Fuzzing forces the race window to be small enough for the kthread_get_run() error path bug and starting threads before updating the local and shared state bug in the stub-up sequence.
--- drivers/usb/usbip/vudc_sysfs.c | 42 +++++++++++++++++++++++++++++++++-------- 1 file changed, 34 insertions(+), 8 deletions(-)
drivers/usb/usbip/vudc_sysfs.c +34 -8
@@ -90,8 +90,9 @@ unlock: } static BIN_ATTR_RO(dev_desc, sizeof(struct usb_device_descriptor)); -static ssize_t usbip_sockfd_store (struct device *dev,struct device_attribute *attr, -const char *in,size_t count) +static ssize_t usbip_sockfd_store (struct device *dev, +struct device_attribute *attr, +const char *in,size_t count) {struct vudc *udc = (struct vudc *)dev_get_drvdata (dev);int rv;
@@ -100,6 +101,8 @@ static ssize_t usbip_sockfd_store(structstruct socket *socket;unsigned long flags;int ret; +struct task_struct *tcp_rx =NULL ; +struct task_struct *tcp_tx =NULL ; rv =kstrtoint (in,0 , &sockfd);if (rv !=0 )
@@ -145,24 +148,47 @@ static ssize_t usbip_sockfd_store(struct goto sock_err; } - udc->ud.tcp_socket = socket; - +/* unlock and create threads and get tasks */ spin_unlock_irq (&udc->ud.lock);spin_unlock_irqrestore (&udc->lock, flags); - udc->ud.tcp_rx =kthread_get_run (&v_rx_loop, - &udc->ud,"vudc_rx" ); - udc->ud.tcp_tx =kthread_get_run (&v_tx_loop, - &udc->ud,"vudc_tx" ); + tcp_rx =kthread_create (&v_rx_loop, &udc->ud,"vudc_rx" ); +if (IS_ERR (tcp_rx)) { +sockfd_put (socket); +return -EINVAL ; + } + tcp_tx =kthread_create (&v_tx_loop, &udc->ud,"vudc_tx" ); +if (IS_ERR (tcp_tx)) { +kthread_stop (tcp_rx); +sockfd_put (socket); +return -EINVAL ; + } + +/* get task structs now */ +get_task_struct (tcp_rx); +get_task_struct (tcp_tx); +/* lock and update udc->ud state */ spin_lock_irqsave (&udc->lock, flags);spin_lock_irq (&udc->ud.lock); + + udc->ud.tcp_socket = socket; + udc->ud.tcp_rx = tcp_rx; + udc->ud.tcp_rx = tcp_tx; udc->ud.status =SDEV_ST_USED ; +spin_unlock_irq (&udc->ud.lock);ktime_get_ts64 (&udc->start_time);v_start_timer (udc); udc->connected =1 ; + +spin_unlock_irqrestore (&udc->lock, flags); + +wake_up_process (udc->ud.tcp_rx); +wake_up_process (udc->ud.tcp_tx); +return count; + }else {if (!udc->connected) {dev_err (dev,"Device not connected" );
[email protected]
maintainerPATCH 125/168 misc/pvpanic: Export module FDT device table
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Shile Zhang <[email protected]>
commit 65527a51c66f4edfa28602643d7dd4fa366eb826 upstream.
Export the module FDT device table to ensure the FDT compatible strings are listed in the module alias. This help the pvpanic driver can be loaded on boot automatically not only the ACPI device, but also the FDT device.
--- drivers/misc/pvpanic.c | 1 + 1 file changed, 1 insertion(+)
drivers/misc/pvpanic.c +1 -0
@@ -166,6 +166,7 @@ static const struct of_device_id pvpanic { .compatible = "qemu,pvpanic-mmio", }, {} }; +MODULE_DEVICE_TABLE (of, pvpanic_mmio_match);static struct platform_driver pvpanic_mmio_driver = { .driver = {
[email protected]
maintainerPATCH 126/168 misc: fastrpc: restrict user apps from sending kernel RPC messages
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Dmitry Baryshkov <[email protected]>
commit 20c40794eb85ea29852d7bc37c55713802a543d6 upstream.
Verify that user applications are not using the kernel RPC message handle to restrict them from directly attaching to guest OS on the remote subsystem. This is a port of CVE-2019-2308 fix.
--- drivers/misc/fastrpc.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/misc/fastrpc.c +5 -0
@@ -924,6 +924,11 @@ static int fastrpc_internal_invoke(strucif (!fl->cctx->rpdev)return -EPIPE ; +if (handle ==FASTRPC_INIT_HANDLE && !kernel) { +dev_warn_ratelimited (fl->sctx->dev,"user app trying to send a kernel RPC message (%d)\n" , handle); +return -EPERM ; + } + ctx =fastrpc_context_alloc (fl, kernel, sc, args);if (IS_ERR (ctx))return PTR_ERR (ctx);
[email protected]
maintainerPATCH 127/168 staging: rtl8192u: fix ->ssid overflow in r8192_wx_set_scan()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Dan Carpenter <[email protected]>
commit 87107518d7a93fec6cdb2559588862afeee800fb upstream.
We need to cap len at IW_ESSID_MAX_SIZE (32) to avoid memory corruption. This can be controlled by the user via the ioctl.
--- drivers/staging/rtl8192u/r8192U_wx.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/staging/rtl8192u/r8192U_wx.c +4 -2
@@ -333,8 +333,10 @@ static int r8192_wx_set_scan(struct net_struct iw_scan_req *req = (struct iw_scan_req *)b;if (req->essid_len) { - ieee->current_network.ssid_len = req->essid_len; -memcpy (ieee->current_network.ssid, req->essid, req->essid_len); +int len =min_t (int, req->essid_len,IW_ESSID_MAX_SIZE ); + + ieee->current_network.ssid_len = len; +memcpy (ieee->current_network.ssid, req->essid, len); } }
[email protected]
maintainerPATCH 128/168 staging: rtl8188eu: prevent ->ssid overflow in rtw_wx_set_scan()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Dan Carpenter <[email protected]>
commit 74b6b20df8cfe90ada777d621b54c32e69e27cd7 upstream.
This code has a check to prevent read overflow but it needs another check to prevent writing beyond the end of the ->ssid[] array.
--- drivers/staging/rtl8188eu/os_dep/ioctl_linux.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/staging/rtl8188eu/os_dep/ioctl_linux.c +4 -2
@@ -1160,9 +1160,11 @@ static int rtw_wx_set_scan(struct net_debreak ; } sec_len = *(pos++); len -=1 ; -if (sec_len >0 && sec_len <= len) { +if (sec_len >0 && + sec_len <= len && + sec_len <=32 ) { ssid[ssid_index].ssid_length = sec_len; -memcpy (ssid[ssid_index].ssid, pos, ssid[ssid_index].ssid_length); +memcpy (ssid[ssid_index].ssid, pos, sec_len); ssid_index++; } pos += sec_len;
[email protected]
maintainerPATCH 129/168 staging: rtl8712: unterminated string leads to read overflow
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Dan Carpenter <[email protected]>
commit d660f4f42ccea50262c6ee90c8e7ad19a69fb225 upstream.
The memdup_user() function does not necessarily return a NUL terminated string so this can lead to a read overflow. Switch from memdup_user() to strndup_user() to fix this bug.
--- drivers/staging/rtl8712/rtl871x_ioctl_linux.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/rtl8712/rtl871x_ioctl_linux.c +1 -1
@@ -928,7 +928,7 @@ static int r871x_wx_set_priv(struct net_struct iw_point *dwrq = (struct iw_point *)awrq; len = dwrq->length; - ext =memdup_user (dwrq->pointer, len); + ext =strndup_user (dwrq->pointer, len);if (IS_ERR (ext))return PTR_ERR (ext);
[email protected]
maintainerPATCH 130/168 staging: rtl8188eu: fix potential memory corruption in rtw_check_beacon_data()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Dan Carpenter <[email protected]>
commit d4ac640322b06095128a5c45ba4a1e80929fe7f3 upstream.
The "ie_len" is a value in the 1-255 range that comes from the user. We have to cap it to ensure that it's not too large or it could lead to memory corruption.
--- drivers/staging/rtl8188eu/core/rtw_ap.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/staging/rtl8188eu/core/rtw_ap.c +5 -0
@@ -784,6 +784,7 @@ int rtw_check_beacon_data(struct adapter/* SSID */ p =rtw_get_ie (ie + _BEACON_IE_OFFSET_, _SSID_IE_, &ie_len, (pbss_network->ie_length - _BEACON_IE_OFFSET_));if (p && ie_len >0 ) { + ie_len =min_t (int, ie_len,sizeof (pbss_network->ssid.ssid));memset (&pbss_network->ssid,0 ,sizeof (struct ndis_802_11_ssid ));memcpy (pbss_network->ssid.ssid, (p +2 ), ie_len); pbss_network->ssid.ssid_length = ie_len;
@@ -802,6 +803,7 @@ int rtw_check_beacon_data(struct adapter/* get supported rates */ p =rtw_get_ie (ie + _BEACON_IE_OFFSET_, _SUPPORTEDRATES_IE_, &ie_len, (pbss_network->ie_length - _BEACON_IE_OFFSET_));if (p) { + ie_len =min_t (int, ie_len,NDIS_802_11_LENGTH_RATES_EX );memcpy (supportRate, p +2 , ie_len); supportRateNum = ie_len; }
@@ -809,6 +811,8 @@ int rtw_check_beacon_data(struct adapter/* get ext_supported rates */ p =rtw_get_ie (ie + _BEACON_IE_OFFSET_, _EXT_SUPPORTEDRATES_IE_, &ie_len, pbss_network->ie_length - _BEACON_IE_OFFSET_);if (p) { + ie_len =min_t (int, ie_len, +NDIS_802_11_LENGTH_RATES_EX - supportRateNum);memcpy (supportRate + supportRateNum, p +2 , ie_len); supportRateNum += ie_len; }
@@ -922,6 +926,7 @@ int rtw_check_beacon_data(struct adapter pht_cap->mcs.rx_mask[0 ] =0xff ; pht_cap->mcs.rx_mask[1 ] =0x0 ; + ie_len =min_t (int, ie_len,sizeof (pmlmepriv->htpriv.ht_cap));memcpy (&pmlmepriv->htpriv.ht_cap, p +2 , ie_len); }
[email protected]
maintainerPATCH 131/168 staging: ks7010: prevent buffer overflow in ks_wlan_set_scan()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Dan Carpenter <[email protected]>
commit e163b9823a0b08c3bb8dc4f5b4b5c221c24ec3e5 upstream.
The user can specify a "req->essid_len" of up to 255 but if it's over IW_ESSID_MAX_SIZE (32) that can lead to memory corruption.
--- drivers/staging/ks7010/ks_wlan_net.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/staging/ks7010/ks_wlan_net.c +4 -2
@@ -1120,6 +1120,7 @@ static int ks_wlan_set_scan(struct net_d {struct ks_wlan_private *priv =netdev_priv (dev);struct iw_scan_req *req =NULL ; +int len;if (priv->sleep_mode ==SLP_SLEEP )return -EPERM ;
@@ -1129,8 +1130,9 @@ static int ks_wlan_set_scan(struct net_dif (wrqu->data.length ==sizeof (struct iw_scan_req ) && wrqu->data.flags &IW_SCAN_THIS_ESSID ) { req = (struct iw_scan_req *)extra; - priv->scan_ssid_len = req->essid_len; -memcpy (priv->scan_ssid, req->essid, priv->scan_ssid_len); + len =min_t (int, req->essid_len,IW_ESSID_MAX_SIZE ); + priv->scan_ssid_len = len; +memcpy (priv->scan_ssid, req->essid, len); }else { priv->scan_ssid_len =0 ; }
[email protected]
maintainerPATCH 132/168 staging: rtl8712: Fix possible buffer overflow in r8712_sitesurvey_cmd
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Lee Gibson <[email protected]>
commit b93c1e3981af19527beee1c10a2bef67a228c48c upstream.
Function r8712_sitesurvey_cmd calls memcpy without checking the length. A user could control that length and trigger a buffer overflow. Fix by checking the length is within the maximum allowed size.
--- drivers/staging/rtl8712/rtl871x_cmd.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/staging/rtl8712/rtl871x_cmd.c +4 -2
@@ -197,8 +197,10 @@ u8 r8712_sitesurvey_cmd(struct _adapter psurveyPara->ss_ssidlen =0 ;memset (psurveyPara->ss_ssid,0 ,IW_ESSID_MAX_SIZE +1 );if ((pssid !=NULL ) && (pssid->SsidLength)) { -memcpy (psurveyPara->ss_ssid, pssid->Ssid, pssid->SsidLength); - psurveyPara->ss_ssidlen =cpu_to_le32 (pssid->SsidLength); +int len =min_t (int, pssid->SsidLength,IW_ESSID_MAX_SIZE ); + +memcpy (psurveyPara->ss_ssid, pssid->Ssid, len); + psurveyPara->ss_ssidlen =cpu_to_le32 (len); }set_fwstate (pmlmepriv, _FW_UNDER_SURVEY);r8712_enqueue_cmd (pcmdpriv, ph2c);
[email protected]
maintainerPATCH 133/168 staging: rtl8192e: Fix possible buffer overflow in _rtl92e_wx_set_scan
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Lee Gibson <[email protected]>
commit 8687bf9ef9551bcf93897e33364d121667b1aadf upstream.
Function _rtl92e_wx_set_scan calls memcpy without checking the length. A user could control that length and trigger a buffer overflow. Fix by checking the length is within the maximum allowed size.
--- drivers/staging/rtl8192e/rtl8192e/rtl_wx.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/staging/rtl8192e/rtl8192e/rtl_wx.c +4 -3
@@ -406,9 +406,10 @@ static int _rtl92e_wx_set_scan(struct nestruct iw_scan_req *req = (struct iw_scan_req *)b;if (req->essid_len) { - ieee->current_network.ssid_len = req->essid_len; -memcpy (ieee->current_network.ssid, req->essid, - req->essid_len); +int len =min_t (int, req->essid_len,IW_ESSID_MAX_SIZE ); + + ieee->current_network.ssid_len = len; +memcpy (ieee->current_network.ssid, req->essid, len); } }
[email protected]
maintainerPATCH 134/168 staging: comedi: addi_apci_1032: Fix endian problem for COS sample
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit 25317f428a78fde71b2bf3f24d05850f08a73a52 upstream.
The Change-Of-State (COS) subdevice supports Comedi asynchronous commands to read 16-bit change-of-state values. However, the interrupt handler is calling comedi_buf_write_samples() with the address of a 32-bit integer &s->state. On bigendian architectures, it will copy 2 bytes from the wrong end of the 32-bit integer. Fix it by transferring the value via a 16-bit integer.
--- drivers/staging/comedi/drivers/addi_apci_1032.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/staging/comedi/drivers/addi_apci_1032.c +3 -1
@@ -260,6 +260,7 @@ static irqreturn_t apci1032_interrupt(instruct apci1032_private *devpriv = dev->private;struct comedi_subdevice *s = dev->read_subdev;unsigned int ctrl; +unsigned short val;/* check interrupt is from this device */ if ((inl(devpriv->amcc_iobase + AMCC_OP_REG_INTCSR) &
@@ -275,7 +276,8 @@ static irqreturn_t apci1032_interrupt(inoutl (ctrl & ~APCI1032_CTRL_INT_ENA , dev->iobase +APCI1032_CTRL_REG ); s->state =inl (dev->iobase +APCI1032_STATUS_REG ) &0xffff ; -comedi_buf_write_samples (s, &s->state,1 ); + val = s->state; +comedi_buf_write_samples (s, &val,1 );comedi_handle_events (dev, s);/* enable the interrupt */
[email protected]
maintainerPATCH 135/168 staging: comedi: addi_apci_1500: Fix endian problem for command sample
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit ac0bbf55ed3be75fde1f8907e91ecd2fd589bde3 upstream.
The digital input subdevice supports Comedi asynchronous commands that read interrupt status information. This uses 16-bit Comedi samples (of which only the bottom 8 bits contain status information). However, the interrupt handler is calling comedi_buf_write_samples() with the address of a 32-bit variable unsigned int status. On a bigendian machine, this will copy 2 bytes from the wrong end of the variable. Fix it by changing the type of the variable to unsigned short.
--- drivers/staging/comedi/drivers/addi_apci_1500.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-)
drivers/staging/comedi/drivers/addi_apci_1500.c +9 -9
@@ -208,7 +208,7 @@ static irqreturn_t apci1500_interrupt(instruct comedi_device *dev = d;struct apci1500_private *devpriv = dev->private;struct comedi_subdevice *s = dev->read_subdev; -unsigned int status =0 ; +unsigned short status =0 ;unsigned int val; val =inl (devpriv->amcc +AMCC_OP_REG_INTCSR );
@@ -238,14 +238,14 @@ static irqreturn_t apci1500_interrupt(in* * Mask Meaning * ---------- ------------------------------------------ -* 0x00000001 Event 1 has occurred -* 0x00000010 Event 2 has occurred -* 0x00000100 Counter/timer 1 has run down (not implemented) -* 0x00001000 Counter/timer 2 has run down (not implemented) -* 0x00010000 Counter 3 has run down (not implemented) -* 0x00100000 Watchdog has run down (not implemented) -* 0x01000000 Voltage error -* 0x10000000 Short-circuit error +* 0b00000001 Event 1 has occurred +* 0b00000010 Event 2 has occurred +* 0b00000100 Counter/timer 1 has run down (not implemented) +* 0b00001000 Counter/timer 2 has run down (not implemented) +* 0b00010000 Counter 3 has run down (not implemented) +* 0b00100000 Watchdog has run down (not implemented) +* 0b01000000 Voltage error +* 0b10000000 Short-circuit error */ comedi_buf_write_samples (s, &status,1 );comedi_handle_events (dev, s);
[email protected]
maintainerPATCH 136/168 staging: comedi: adv_pci1710: Fix endian problem for AI command data
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit b2e78630f733a76508b53ba680528ca39c890e82 upstream.
The analog input subdevice supports Comedi asynchronous commands that use Comedi's 16-bit sample format. However, the calls to comedi_buf_write_samples() are passing the address of a 32-bit integer variable. On bigendian machines, this will copy 2 bytes from the wrong end of the 32-bit value. Fix it by changing the type of the variables holding the sample value to unsigned short. The type of the val parameter of pci1710_ai_read_sample() is changed to unsigned short * accordingly. The type of the val variable in pci1710_ai_insn_read() is also changed to unsigned short since its address is passed to pci1710_ai_read_sample().
--- drivers/staging/comedi/drivers/adv_pci1710.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/staging/comedi/drivers/adv_pci1710.c +5 -5
@@ -300,11 +300,11 @@ static int pci1710_ai_eoc(struct comedi_static int pci1710_ai_read_sample (struct comedi_device *dev,struct comedi_subdevice *s,unsigned int cur_chan, -unsigned int *val) +unsigned short *val) {const struct boardtype *board = dev->board_ptr;struct pci1710_private *devpriv = dev->private; -unsigned int sample; +unsigned short sample;unsigned int chan; sample =inw (dev->iobase +PCI171X_AD_DATA_REG );
@@ -345,7 +345,7 @@ static int pci1710_ai_insn_read(struct cpci1710_ai_setup_chanlist (dev, s, &insn->chanspec,1 ,1 );for (i =0 ; i < insn->n; i++) { -unsigned int val; +unsigned short val;/* start conversion */ outw (0 , dev->iobase +PCI171X_SOFTTRG_REG );
@@ -395,7 +395,7 @@ static void pci1710_handle_every_sample( {struct comedi_cmd *cmd = &s->async->cmd;unsigned int status; -unsigned int val; +unsigned short val;int ret; status =inw (dev->iobase +PCI171X_STATUS_REG );
@@ -455,7 +455,7 @@ static void pci1710_handle_fifo(struct c }for (i =0 ; i < devpriv->max_samples; i++) { -unsigned int val; +unsigned short val;int ret; ret =pci1710_ai_read_sample (dev, s, s->async->cur_chan, &val);
[email protected]
maintainerPATCH 137/168 staging: comedi: das6402: Fix endian problem for AI command data
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit 1c0f20b78781b9ca50dc3ecfd396d0db5b141890 upstream.
The analog input subdevice supports Comedi asynchronous commands that use Comedi's 16-bit sample format. However, the call to comedi_buf_write_samples() is passing the address of a 32-bit integer variable. On bigendian machines, this will copy 2 bytes from the wrong end of the 32-bit value. Fix it by changing the type of the variable holding the sample value to unsigned short.
--- drivers/staging/comedi/drivers/das6402.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/comedi/drivers/das6402.c +1 -1
@@ -186,7 +186,7 @@ static irqreturn_t das6402_interrupt(intif (status &DAS6402_STATUS_FFULL ) { async->events |=COMEDI_CB_OVERFLOW ; }else if (status &DAS6402_STATUS_FFNE ) { -unsigned int val; +unsigned short val; val =das6402_ai_read_sample (dev, s);comedi_buf_write_samples (s, &val,1 );
[email protected]
maintainerPATCH 138/168 staging: comedi: das800: Fix endian problem for AI command data
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit 459b1e8c8fe97fcba0bd1b623471713dce2c5eaf upstream.
The analog input subdevice supports Comedi asynchronous commands that use Comedi's 16-bit sample format. However, the call to comedi_buf_write_samples() is passing the address of a 32-bit integer variable. On bigendian machines, this will copy 2 bytes from the wrong end of the 32-bit value. Fix it by changing the type of the variable holding the sample value to unsigned short.
--- drivers/staging/comedi/drivers/das800.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/comedi/drivers/das800.c +1 -1
@@ -427,7 +427,7 @@ static irqreturn_t das800_interrupt(intstruct comedi_cmd *cmd;unsigned long irq_flags;unsigned int status; -unsigned int val; +unsigned short val;bool fifo_empty;bool fifo_overflow;int i;
[email protected]
maintainerPATCH 139/168 staging: comedi: dmm32at: Fix endian problem for AI command data
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit 54999c0d94b3c26625f896f8e3460bc029821578 upstream.
The analog input subdevice supports Comedi asynchronous commands that use Comedi's 16-bit sample format. However, the call to comedi_buf_write_samples() is passing the address of a 32-bit integer variable. On bigendian machines, this will copy 2 bytes from the wrong end of the 32-bit value. Fix it by changing the type of the variable holding the sample value to unsigned short.
[Note: the bug was introduced in commit 1700529b24cc ("staging: comedi:
dmm32at: use comedi_buf_write_samples()") but the patch applies better to the later (but in the same kernel release) commit 0c0eadadcbe6e ("staging: comedi: dmm32at: introduce dmm32_ai_get_sample()").]
--- drivers/staging/comedi/drivers/dmm32at.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/comedi/drivers/dmm32at.c +1 -1
@@ -404,7 +404,7 @@ static irqreturn_t dmm32at_isr(int irq, {struct comedi_device *dev = d;unsigned char intstat; -unsigned int val; +unsigned short val;int i;if (!dev->attached) {
[email protected]
maintainerPATCH 140/168 staging: comedi: me4000: Fix endian problem for AI command data
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit b39dfcced399d31e7c4b7341693b18e01c8f655e upstream.
The analog input subdevice supports Comedi asynchronous commands that use Comedi's 16-bit sample format. However, the calls to comedi_buf_write_samples() are passing the address of a 32-bit integer variable. On bigendian machines, this will copy 2 bytes from the wrong end of the 32-bit value. Fix it by changing the type of the variable holding the sample value to unsigned short.
--- drivers/staging/comedi/drivers/me4000.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/comedi/drivers/me4000.c +1 -1
@@ -924,7 +924,7 @@ static irqreturn_t me4000_ai_isr(int irqstruct comedi_subdevice *s = dev->read_subdev;int i;int c =0 ; -unsigned int lval; +unsigned short lval;if (!dev->attached)return IRQ_NONE ;
[email protected]
maintainerPATCH 141/168 staging: comedi: pcl711: Fix endian problem for AI command data
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit a084303a645896e834883f2c5170d044410dfdb3 upstream.
The analog input subdevice supports Comedi asynchronous commands that use Comedi's 16-bit sample format. However, the call to comedi_buf_write_samples() is passing the address of a 32-bit integer variable. On bigendian machines, this will copy 2 bytes from the wrong end of the 32-bit value. Fix it by changing the type of the variable holding the sample value to unsigned short.
--- drivers/staging/comedi/drivers/pcl711.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/comedi/drivers/pcl711.c +1 -1
@@ -184,7 +184,7 @@ static irqreturn_t pcl711_interrupt(intstruct comedi_device *dev = d;struct comedi_subdevice *s = dev->read_subdev;struct comedi_cmd *cmd = &s->async->cmd; -unsigned int data; +unsigned short data;if (!dev->attached) {dev_err (dev->class_dev,"spurious interrupt\n" );
[email protected]
maintainerPATCH 142/168 staging: comedi: pcl818: Fix endian problem for AI command data
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ian Abbott <[email protected]>
commit 148e34fd33d53740642db523724226de14ee5281 upstream.
The analog input subdevice supports Comedi asynchronous commands that use Comedi's 16-bit sample format. However, the call to comedi_buf_write_samples() is passing the address of a 32-bit integer parameter. On bigendian machines, this will copy 2 bytes from the wrong end of the 32-bit value. Fix it by changing the type of the parameter holding the sample value to unsigned short.
[Note: the bug was introduced in commit edf4537bcbf5 ("staging: comedi:
pcl818: use comedi_buf_write_samples()") but the patch applies better to commit d615416de615 ("staging: comedi: pcl818: introduce pcl818_ai_write_sample()").]
--- drivers/staging/comedi/drivers/pcl818.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/comedi/drivers/pcl818.c +1 -1
@@ -423,7 +423,7 @@ static int pcl818_ai_eoc(struct comedi_dstatic bool pcl818_ai_write_sample (struct comedi_device *dev,struct comedi_subdevice *s, -unsigned int chan,unsigned int val) +unsigned int chan,unsigned short val) {struct pcl818_private *devpriv = dev->private;struct comedi_cmd *cmd = &s->async->cmd;
[email protected]
maintainerPATCH 143/168 sh_eth: fix TRSCER mask for R7S72100
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Sergey Shtylyov <[email protected]>
[ Upstream commit 75be7fb7f978202c4c3a1a713af4485afb2ff5f6 ]
According to the RZ/A1H Group, RZ/A1M Group User's Manual: Hardware, Rev. 4.00, the TRSCER register has bit 9 reserved, hence we can't use the driver's default TRSCER mask. Add the explicit initializer for sh_eth_cpu_data::trscer_err_mask for R7S72100.
--- drivers/net/ethernet/renesas/sh_eth.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/renesas/sh_eth.c +2 -0
@@ -610,6 +610,8 @@ static struct sh_eth_cpu_data r7s72100_data = { EESR_TDE, .fdr_value = 0x0000070f, + .trscer_err_mask = DESC_I_RINT8 | DESC_I_RINT5, + .no_psr = 1, .apr = 1, .mpr = 1,
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 144/168 arm64/mm: Fix pfn_valid() for ZONE_DEVICE based memory
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Anshuman Khandual <[email protected]>
[ Upstream commit eeb0753ba27b26f609e61f9950b14f1b934fe429 ]
pfn_valid() validates a pfn but basically it checks for a valid struct page backing for that pfn. It should always return positive for memory ranges backed with struct page mapping. But currently pfn_valid() fails for all ZONE_DEVICE based memory types even though they have struct page mapping.
pfn_valid() asserts that there is a memblock entry for a given pfn without MEMBLOCK_NOMAP flag being set. The problem with ZONE_DEVICE based memory is that they do not have memblock entries. Hence memblock_is_map_memory() will invariably fail via memblock_search() for a ZONE_DEVICE based address. This eventually fails pfn_valid() which is wrong. memblock_is_map_memory() needs to be skipped for such memory ranges. As ZONE_DEVICE memory gets hotplugged into the system via memremap_pages() called from a driver, their respective memory sections will not have SECTION_IS_EARLY set.
Normal hotplug memory will never have MEMBLOCK_NOMAP set in their memblock regions. Because the flag MEMBLOCK_NOMAP was specifically designed and set for firmware reserved memory regions. memblock_is_map_memory() can just be skipped as its always going to be positive and that will be an optimization for the normal hotplug memory. Like ZONE_DEVICE based memory, all normal hotplugged memory too will not have SECTION_IS_EARLY set for their sections
Skipping memblock_is_map_memory() for all non early memory sections would fix pfn_valid() problem for ZONE_DEVICE based memory and also improve its performance for normal hotplug memory as well.
--- arch/arm64/mm/init.c | 12 ++++++++++++ 1 file changed, 12 insertions(+)
arch/arm64/mm/init.c +12 -0
@@ -245,6 +245,18 @@ int pfn_valid(unsigned long pfn)if (!valid_section (__nr_to_section (pfn_to_section_nr (pfn))))return 0 ; + +/* +* ZONE_DEVICE memory does not have the memblock entries. +* memblock_is_map_memory() check for ZONE_DEVICE based +* addresses will always fail. Even the normal hotplugged +* memory will never have MEMBLOCK_NOMAP flag set in their +* memblock entries. Skip memblock search for all non early +* memory sections covering all of hotplug memory including +* both normal and ZONE_DEVICE based. +*/ +if (!early_section (__pfn_to_section (pfn))) +return pfn_section_valid (__pfn_to_section (pfn), pfn);#endif return memblock_is_map_memory (addr); }
Signature · 9 lines
2.30.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
[email protected]
maintainerPATCH 145/168 SUNRPC: Set memalloc_nofs_save() for sync tasks
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Benjamin Coddington <[email protected]>
[ Upstream commit f0940f4b3284a00f38a5d42e6067c2aaa20e1f2e ]
We could recurse into NFS doing memory reclaim while sending a sync task, which might result in a deadlock. Set memalloc_nofs_save for sync task execution.
--- net/sunrpc/sched.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/sunrpc/sched.c +4 -1
@@ -990,8 +990,11 @@ void rpc_execute(struct rpc_task *task)rpc_set_active (task);rpc_make_runnable (rpciod_workqueue, task); -if (!is_async) +if (!is_async) { +unsigned int pflags =memalloc_nofs_save ();__rpc_execute (task); +memalloc_nofs_restore (pflags); + } }static void rpc_async_schedule (struct work_struct *work)
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 146/168 NFS: Dont revalidate the directory permissions on a lookup failure
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Trond Myklebust <[email protected]>
[ Upstream commit 82e7ca1334ab16e2e04fafded1cab9dfcdc11b40 ]
There should be no reason to expect the directory permissions to change just because the directory contents changed or a negative lookup timed out. So let's avoid doing a full call to nfs_mark_for_revalidate() in that case.
Furthermore, if this is a negative dentry, and we haven't actually done a new lookup, then we have no reason yet to believe the directory has changed at all. So let's remove the gratuitous directory inode invalidation altogether when called from nfs_lookup_revalidate_negative().
--- fs/nfs/dir.c | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-)
fs/nfs/dir.c +17 -3
@@ -1073,6 +1073,15 @@ int nfs_lookup_verify_inode(struct inode *inode, unsigned int flags) goto out; } +static void nfs_mark_dir_for_revalidate (struct inode *inode) +{ +struct nfs_inode *nfsi =NFS_I (inode); + +spin_lock (&inode->i_lock); + nfsi->cache_validity |=NFS_INO_REVAL_PAGECACHE ; +spin_unlock (&inode->i_lock); +} +/* * We judge how long we want to trust negative * dentries by looking at the parent inode mtime.
@@ -1107,7 +1116,6 @@ nfs_lookup_revalidate_done(struct inode *dir, struct dentry *dentry, __func__, dentry);return 1 ;case 0 : -nfs_mark_for_revalidate (dir);if (inode &&S_ISDIR (inode->i_mode)) {/* Purge readdir caches. */ nfs_zap_caches (inode);
@@ -1188,6 +1196,13 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry,nfs_free_fattr (fattr);nfs_free_fhandle (fhandle);nfs4_label_free (label); + +/* +* If the lookup failed despite the dentry change attribute being +* a match, then we should revalidate the directory cache. +*/ +if (!ret &&nfs_verify_change_attribute (dir, dentry->d_time)) +nfs_mark_dir_for_revalidate (dir);return nfs_lookup_revalidate_done (dir, dentry, inode, ret); }
@@ -1230,7 +1245,7 @@ nfs_do_lookup_revalidate(struct inode *dir, struct dentry *dentry, error =nfs_lookup_verify_inode (inode, flags);if (error) {if (error == -ESTALE ) -nfs_zap_caches (dir); +nfs_mark_dir_for_revalidate (dir); goto out_bad; }nfs_advise_use_readdirplus (dir);
@@ -1725,7 +1740,6 @@ nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle,dput (parent);return d; out_error: -nfs_mark_for_revalidate (dir); d =ERR_PTR (error); goto out; }
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 147/168 NFS: Dont gratuitously clear the inode cache when lookup failed
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Trond Myklebust <[email protected]>
[ Upstream commit 47397915ede0192235474b145ebcd81b37b03624 ]
The fact that the lookup revalidation failed, does not mean that the inode contents have changed.
--- fs/nfs/dir.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-)
fs/nfs/dir.c +8 -12
@@ -1116,18 +1116,14 @@ nfs_lookup_revalidate_done(struct inode *dir, struct dentry *dentry, __func__, dentry);return 1 ;case 0 : -if (inode &&S_ISDIR (inode->i_mode)) { -/* Purge readdir caches. */ -nfs_zap_caches (inode); -/* -* We can't d_drop the root of a disconnected tree: -* its d_hash is on the s_anon list and d_drop() would hide -* it from shrink_dcache_for_unmount(), leading to busy -* inodes on unmount and further oopses. -*/ -if (IS_ROOT (dentry)) -return 1 ; - } +/* +* We can't d_drop the root of a disconnected tree: +* its d_hash is on the s_anon list and d_drop() would hide +* it from shrink_dcache_for_unmount(), leading to busy +* inodes on unmount and further oopses. +*/ +if (inode &&IS_ROOT (dentry)) +return 1 ;dfprintk (LOOKUPCACHE ,"NFS: %s(%pd2) is invalid\n" , __func__, dentry);return 0 ;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 148/168 NFSv4.2: fix return value of _nfs4_get_security_label()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ondrej Mosnacek <[email protected]>
[ Upstream commit 53cb245454df5b13d7063162afd7a785aed6ebf2 ]
An xattr 'get' handler is expected to return the length of the value on success, yet _nfs4_get_security_label() (and consequently also nfs4_xattr_get_nfs4_label(), which is used as an xattr handler) returns just 0 on success.
Fix this by returning label.len instead, which contains the length of the result.
--- fs/nfs/nfs4proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/nfs/nfs4proc.c +1 -1
@@ -5830,7 +5830,7 @@ static int _nfs4_get_security_label(struct inode *inode, void *buf,return ret;if (!(fattr.valid &NFS_ATTR_FATTR_V4_SECURITY_LABEL ))return -ENOENT ; -return 0 ; +return label.len; } static int nfs4_get_security_label(struct inode *inode, void *buf,
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 149/168 block: rsxx: fix error return code of rsxx_pci_probe()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Jia-Ju Bai <[email protected]>
[ Upstream commit df66617bfe87487190a60783d26175b65d2502ce ]
When create_singlethread_workqueue returns NULL to card->event_wq, no error return code of rsxx_pci_probe() is assigned.
To fix this bug, st is assigned with -ENOMEM in this case.
--- drivers/block/rsxx/core.c | 1 + 1 file changed, 1 insertion(+)
drivers/block/rsxx/core.c +1 -0
@@ -869,6 +869,7 @@ static int rsxx_pci_probe(struct pci_dev *dev, card->event_wq =create_singlethread_workqueue (DRIVER_NAME "_event" );if (!card->event_wq) {dev_err (CARD_TO_DEV (card),"Failed card event setup.\n" ); + st = -ENOMEM ; goto failed_event_handler; }
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 150/168 configfs: fix a use-after-free in __configfs_open_file
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Daiyue Zhang <[email protected]>
[ Upstream commit 14fbbc8297728e880070f7b077b3301a8c698ef9 ]
Commit b0841eefd969 ("configfs: provide exclusion between IO and removals") uses ->frag_dead to mark the fragment state, thus no bothering with extra refcount on config_item when opening a file. The configfs_get_config_item was removed in __configfs_open_file, but not with config_item_put. So the refcount on config_item will lost its balance, causing use-after-free issues in some occasions like this:
Test:
1. Mount configfs on /config with read-only items:
drwxrwx--- 289 root root 0 2021-04-01 11:55 /config drwxr-xr-x 2 root root 0 2021-04-01 11:54 /config/a --w--w--w- 1 root root 4096 2021-04-01 11:53 /config/a/1.txt
......
2. Then run:
for file in /config
do
echo $file
grep -R 'key' $file
done
3. __configfs_open_file will be called in parallel, the first one got called will do:
if (file->f_mode & FMODE_READ) {
if (!(inode->i_mode & S_IRUGO))
goto out_put_module;
config_item_put(buffer->item);
kref_put()
package_details_release()
kfree()the other one will run into use-after-free issues like this:
BUG: KASAN: use-after-free in __configfs_open_file+0x1bc/0x3b0 Read of size 8 at addr fffffff155f02480 by task grep/13096 CPU: 0 PID: 13096 Comm: grep VIP: 00 Tainted: G W 4.14.116-kasan #1 TGID: 13096 Comm: grep Call trace: dump_stack+0x118/0x160 kasan_report+0x22c/0x294 __asan_load8+0x80/0x88 __configfs_open_file+0x1bc/0x3b0 configfs_open_file+0x28/0x34 do_dentry_open+0x2cc/0x5c0 vfs_open+0x80/0xe0 path_openat+0xd8c/0x2988 do_filp_open+0x1c4/0x2fc do_sys_open+0x23c/0x404 SyS_openat+0x38/0x48 Allocated by task 2138: kasan_kmalloc+0xe0/0x1ac kmem_cache_alloc_trace+0x334/0x394 packages_make_item+0x4c/0x180 configfs_mkdir+0x358/0x740 vfs_mkdir2+0x1bc/0x2e8 SyS_mkdirat+0x154/0x23c el0_svc_naked+0x34/0x38 Freed by task 13096: kasan_slab_free+0xb8/0x194 kfree+0x13c/0x910 package_details_release+0x524/0x56c kref_put+0xc4/0x104 config_item_put+0x24/0x34 __configfs_open_file+0x35c/0x3b0 configfs_open_file+0x28/0x34 do_dentry_open+0x2cc/0x5c0 vfs_open+0x80/0xe0 path_openat+0xd8c/0x2988 do_filp_open+0x1c4/0x2fc do_sys_open+0x23c/0x404 SyS_openat+0x38/0x48 el0_svc_naked+0x34/0x38
To fix this issue, remove the config_item_put in __configfs_open_file to balance the refcount of config_item.
--- fs/configfs/file.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
fs/configfs/file.c +2 -4
@@ -378,7 +378,7 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type attr =to_attr (dentry);if (!attr) - goto out_put_item; + goto out_free_buffer;if (type &CONFIGFS_ITEM_BIN_ATTR ) { buffer->bin_attr =to_bin_attr (dentry);
@@ -391,7 +391,7 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type/* Grab the module reference for this attribute if we have one */ error = -ENODEV ;if (!try_module_get (buffer->owner)) - goto out_put_item; + goto out_free_buffer; error = -EACCES ;if (!buffer->item->ci_type)
@@ -435,8 +435,6 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type out_put_module:module_put (buffer->owner); -out_put_item: -config_item_put (buffer->item); out_free_buffer:up_read (&frag->frag_sem);kfree (buffer);
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 151/168 arm64: mm: use a 48-bit ID map when possible on 52-bit VA builds
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Ard Biesheuvel <[email protected]>
[ Upstream commit 7ba8f2b2d652cd8d8a2ab61f4be66973e70f9f88 ]
52-bit VA kernels can run on hardware that is only 48-bit capable, but configure the ID map as 52-bit by default. This was not a problem until recently, because the special T0SZ value for a 52-bit VA space was never programmed into the TCR register anwyay, and because a 52-bit ID map happens to use the same number of translation levels as a 48-bit one.
This behavior was changed by commit 1401bef703a4 ("arm64: mm: Always update TCR_EL1 from __cpu_set_tcr_t0sz()"), which causes the unsupported T0SZ value for a 52-bit VA to be programmed into TCR_EL1. While some hardware simply ignores this, Mark reports that Amberwing systems choke on this, resulting in a broken boot. But even before that commit, the unsupported idmap_t0sz value was exposed to KVM and used to program TCR_EL2 incorrectly as well.
Given that we already have to deal with address spaces being either 48-bit or 52-bit in size, the cleanest approach seems to be to simply default to a 48-bit VA ID map, and only switch to a 52-bit one if the placement of the kernel in DRAM requires it. This is guaranteed not to happen unless the system is actually 52-bit VA capable.
--- arch/arm64/include/asm/mmu_context.h | 5 +---- arch/arm64/kernel/head.S | 2 +- arch/arm64/mm/mmu.c | 2 +- 3 files changed, 3 insertions(+), 6 deletions(-)
arch/arm64/include/asm/mmu_context.h +1 -4
@@ -63,10 +63,7 @@ extern u64 idmap_ptrs_per_pgd;static inline bool __cpu_uses_extended_idmap (void ) { -if (IS_ENABLED (CONFIG_ARM64_VA_BITS_52 )) -return false; - -return unlikely (idmap_t0sz !=TCR_T0SZ (VA_BITS )); +return unlikely (idmap_t0sz !=TCR_T0SZ (vabits_actual)); }/*
arch/arm64/kernel/head.S +1 -1
@@ -337,7 +337,7 @@ __create_page_tables: */ adrp x5, __idmap_text_end clz x5, x5 - cmp x5, TCR_T0SZ(VA_BITS) // default T0SZ small enough? + cmp x5, TCR_T0SZ(VA_BITS_MIN) // default T0SZ small enough? b.ge 1f // .. then skip VA range extension adr_l x6, idmap_t0sz
arch/arm64/mm/mmu.c +1 -1
@@ -38,7 +38,7 @@#define NO_BLOCK_MAPPINGS BIT(0)#define NO_CONT_MAPPINGS BIT(1) -u64 idmap_t0sz =TCR_T0SZ (VA_BITS ); +u64 idmap_t0sz =TCR_T0SZ (VA_BITS_MIN );u64 idmap_ptrs_per_pgd =PTRS_PER_PGD ; u64 __section(".mmuoff.data.write") vabits_actual;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 152/168 hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Anna-Maria Behnsen <[email protected]>
[ Upstream commit 46eb1701c046cc18c032fa68f3c8ccbf24483ee4 ]
hrtimer_force_reprogram() and hrtimer_interrupt() invokes __hrtimer_get_next_event() to find the earliest expiry time of hrtimer bases. __hrtimer_get_next_event() does not update cpu_base::[softirq_]_expires_next to preserve reprogramming logic. That needs to be done at the callsites.
hrtimer_force_reprogram() updates cpu_base::softirq_expires_next only when the first expiring timer is a softirq timer and the soft interrupt is not activated. That's wrong because cpu_base::softirq_expires_next is left stale when the first expiring timer of all bases is a timer which expires in hard interrupt context. hrtimer_interrupt() does never update cpu_base::softirq_expires_next which is wrong too.
That becomes a problem when clock_settime() sets CLOCK_REALTIME forward and the first soft expiring timer is in the CLOCK_REALTIME_SOFT base. Setting CLOCK_REALTIME forward moves the clock MONOTONIC based expiry time of that timer before the stale cpu_base::softirq_expires_next.
cpu_base::softirq_expires_next is cached to make the check for raising the soft interrupt fast. In the above case the soft interrupt won't be raised until clock monotonic reaches the stale cpu_base::softirq_expires_next value. That's incorrect, but what's worse it that if the softirq timer becomes the first expiring timer of all clock bases after the hard expiry timer has been handled the reprogramming of the clockevent from hrtimer_interrupt() will result in an interrupt storm. That happens because the reprogramming does not use cpu_base::softirq_expires_next, it uses __hrtimer_get_next_event() which returns the actual expiry time. Once clock MONOTONIC reaches cpu_base::softirq_expires_next the soft interrupt is raised and the storm subsides.
Change the logic in hrtimer_force_reprogram() to evaluate the soft and hard bases seperately, update softirq_expires_next and handle the case when a soft expiring timer is the first of all bases by comparing the expiry times and updating the required cpu base fields. Split this functionality into a separate function to be able to use it in hrtimer_interrupt() as well without copy paste.
--- kernel/time/hrtimer.c | 60 ++++++++++++++++++++++++++++--------------- 1 file changed, 39 insertions(+), 21 deletions(-)
kernel/time/hrtimer.c +39 -21
@@ -547,8 +547,11 @@ static ktime_t __hrtimer_next_event_base(struct hrtimer_cpu_base *cpu_base, }/* -* Recomputes cpu_base::*next_timer and returns the earliest expires_next but -* does not set cpu_base::*expires_next, that is done by hrtimer_reprogram. +* Recomputes cpu_base::*next_timer and returns the earliest expires_next +* but does not set cpu_base::*expires_next, that is done by +* hrtimer[_force]_reprogram and hrtimer_interrupt only. When updating +* cpu_base::*expires_next right away, reprogramming logic would no longer +* work. * * When a softirq is pending, we can ignore the HRTIMER_ACTIVE_SOFT bases, * those timers will get run whenever the softirq gets handled, at the end of
@@ -589,6 +592,37 @@ __hrtimer_get_next_event(struct hrtimer_cpu_base *cpu_base, unsigned int active_return expires_next; } +static ktime_t hrtimer_update_next_event (struct hrtimer_cpu_base *cpu_base) +{ +ktime_t expires_next, soft =KTIME_MAX ; + +/* +* If the soft interrupt has already been activated, ignore the +* soft bases. They will be handled in the already raised soft +* interrupt. +*/ +if (!cpu_base->softirq_activated) { + soft =__hrtimer_get_next_event (cpu_base,HRTIMER_ACTIVE_SOFT ); +/* +* Update the soft expiry time. clock_settime() might have +* affected it. +*/ + cpu_base->softirq_expires_next = soft; + } + + expires_next =__hrtimer_get_next_event (cpu_base,HRTIMER_ACTIVE_HARD ); +/* +* If a softirq timer is expiring first, update cpu_base->next_timer +* and program the hardware with the soft expiry time. +*/ +if (expires_next > soft) { + cpu_base->next_timer = cpu_base->softirq_next_timer; + expires_next = soft; + } + +return expires_next; +} +static inline ktime_t hrtimer_update_base (struct hrtimer_cpu_base *base) {ktime_t *offs_real = &base->clock_base[HRTIMER_BASE_REALTIME ].offset;
@@ -629,23 +663,7 @@ hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) {ktime_t expires_next; -/* -* Find the current next expiration time. -*/ - expires_next =__hrtimer_get_next_event (cpu_base,HRTIMER_ACTIVE_ALL ); - -if (cpu_base->next_timer && cpu_base->next_timer->is_soft) { -/* -* When the softirq is activated, hrtimer has to be -* programmed with the first hard hrtimer because soft -* timer interrupt could occur too late. -*/ -if (cpu_base->softirq_activated) - expires_next =__hrtimer_get_next_event (cpu_base, -HRTIMER_ACTIVE_HARD ); -else - cpu_base->softirq_expires_next = expires_next; - } + expires_next =hrtimer_update_next_event (cpu_base);if (skip_equal && expires_next == cpu_base->expires_next)return ;
@@ -1640,8 +1658,8 @@ void hrtimer_interrupt(struct clock_event_device *dev)__hrtimer_run_queues (cpu_base, now, flags,HRTIMER_ACTIVE_HARD ); -/* Reevaluate the clock bases for the next expiry */ - expires_next =__hrtimer_get_next_event (cpu_base,HRTIMER_ACTIVE_ALL ); +/* Reevaluate the clock bases for the [soft] next expiry */ + expires_next =hrtimer_update_next_event (cpu_base);/* * Store the new expiry value so the migration code can verify * against it.
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 153/168 stop_machine: mark helpers __always_inline
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Arnd Bergmann <[email protected]>
[ Upstream commit cbf78d85079cee662c45749ef4f744d41be85d48 ]
With clang-13, some functions only get partially inlined, with a specialized version referring to a global variable. This triggers a harmless build-time check for the intel-rng driver:
WARNING: modpost: drivers/char/hw_random/intel-rng.o(.text+0xe): Section mismatch in reference from the function stop_machine() to the function .init.text:intel_rng_hw_init() The function stop_machine() references
the function __init intel_rng_hw_init().
This is often because stop_machine lacks a __init
annotation or the annotation of intel_rng_hw_init is wrong.
In this instance, an easy workaround is to force the stop_machine() function to be inline, along with related interfaces that did not show the same behavior at the moment, but theoretically could.
The combination of the two patches listed below triggers the behavior in clang-13, but individually these commits are correct.
--- include/linux/stop_machine.h | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
include/linux/stop_machine.h +6 -5
@@ -139,7 +139,7 @@ int stop_machine_from_inactive_cpu(cpu_stop_fn_t fn, void *data, const struct cpumask *cpus);#else /* CONFIG_SMP || CONFIG_HOTPLUG_CPU */ -static inline int stop_machine_cpuslocked (cpu_stop_fn_t fn,void *data, +static __always_inline int stop_machine_cpuslocked(cpu_stop_fn_t fn, void *data,const struct cpumask *cpus) {unsigned long flags;
@@ -150,14 +150,15 @@ static inline int stop_machine_cpuslocked(cpu_stop_fn_t fn, void *data,return ret; } -static inline int stop_machine (cpu_stop_fn_t fn,void *data, -const struct cpumask *cpus) +static __always_inline int +stop_machine (cpu_stop_fn_t fn,void *data,const struct cpumask *cpus) {return stop_machine_cpuslocked (fn, data, cpus); } -static inline int stop_machine_from_inactive_cpu (cpu_stop_fn_t fn,void *data, -const struct cpumask *cpus) +static __always_inline int +stop_machine_from_inactive_cpu (cpu_stop_fn_t fn,void *data, +const struct cpumask *cpus) {return stop_machine (fn, data, cpus); }
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 154/168 include/linux/sched/mm.h: use rcu_dereference in in_vfork()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Matthew Wilcox (Oracle) <[email protected]>
[ Upstream commit 149fc787353f65b7e72e05e7b75d34863266c3e2 ]
Fix a sparse warning by using rcu_dereference(). Technically this is a bug and a sufficiently aggressive compiler could reload the `real_parent' pointer outside the protection of the rcu lock (and access freed memory), but I think it's pretty unlikely to happen.
--- include/linux/sched/mm.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
include/linux/sched/mm.h +2 -1
@@ -167,7 +167,8 @@ static inline bool in_vfork(struct task_struct *tsk)* another oom-unkillable task does this it should blame itself. */ rcu_read_lock (); - ret = tsk->vfork_done && tsk->real_parent->mm == tsk->mm; + ret = tsk->vfork_done && +rcu_dereference (tsk->real_parent)->mm == tsk->mm;rcu_read_unlock ();return ret;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 155/168 prctl: fix PR_SET_MM_AUXV kernel stack leak
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Alexey Dobriyan <[email protected]>
[ Upstream commit c995f12ad8842dbf5cfed113fb52cdd083f5afd1 ]
Doing a
prctl(PR_SET_MM, PR_SET_MM_AUXV, addr, 1);
will copy 1 byte from userspace to (quite big) on-stack array and then stash everything to mm->saved_auxv. AT_NULL terminator will be inserted at the very end.
/proc/*/auxv handler will find that AT_NULL terminator and copy original stack contents to userspace.
This devious scheme requires CAP_SYS_RESOURCE.
--- kernel/sys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/sys.c +1 -1
@@ -2062,7 +2062,7 @@ static int prctl_set_auxv(struct mm_struct *mm, unsigned long addr,* up to the caller to provide sane values here, otherwise userspace * tools which use this vector might be unhappy. */ -unsigned long user_auxv[AT_VECTOR_SIZE ]; +unsigned long user_auxv[AT_VECTOR_SIZE ] = {};if (len >sizeof (user_auxv))return -EINVAL ;
Signature · 1 line
2.30.1
[email protected]
maintainerPATCH 156/168 zram: fix return value on writeback_store
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Minchan Kim <[email protected]>
commit 57e0076e6575a7b7cef620a0bd2ee2549ef77818 upstream.
writeback_store's return value is overwritten by submit_bio_wait's return value. Thus, writeback_store will return zero since there was no IO error. In the end, write syscall from userspace will see the zero as return value, which could make the process stall to keep trying the write until it will succeed.
--- drivers/block/zram/zram_drv.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)
drivers/block/zram/zram_drv.c +8 -3
@@ -627,7 +627,7 @@ static ssize_t writeback_store(struct destruct bio_vec bio_vec;struct page *page;ssize_t ret = len; -int mode; +int mode, err;unsigned long blk_idx =0 ;if (sysfs_streq (buf,"idle" ))
@@ -719,12 +719,17 @@ static ssize_t writeback_store(struct de* XXX: A single page IO would be inefficient for write * but it would be not bad as starter. */ - ret =submit_bio_wait (&bio); -if (ret) { + err =submit_bio_wait (&bio); +if (err) {zram_slot_lock (zram, index);zram_clear_flag (zram, index,ZRAM_UNDER_WB );zram_clear_flag (zram, index,ZRAM_IDLE );zram_slot_unlock (zram, index); +/* +* Return last IO error unless every IO were +* not suceeded. +*/ + ret = err;continue ; }
[email protected]
maintainerPATCH 157/168 sched/membarrier: fix missing local execution of ipi_sync_rq_state()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Mathieu Desnoyers <[email protected]>
commit ce29ddc47b91f97e7f69a0fb7cbb5845f52a9825 upstream.
The function sync_runqueues_membarrier_state() should copy the membarrier state from the @mm received as parameter to each runqueue currently running tasks using that mm.
However, the use of smp_call_function_many() skips the current runqueue, which is unintended. Replace by a call to on_each_cpu_mask().
--- kernel/sched/membarrier.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
kernel/sched/membarrier.c +1 -3
@@ -265,9 +265,7 @@ static int sync_runqueues_membarrier_sta }rcu_read_unlock (); -preempt_disable (); -smp_call_function_many (tmpmask, ipi_sync_rq_state, mm,1 ); -preempt_enable (); +on_each_cpu_mask (tmpmask, ipi_sync_rq_state, mm, true);free_cpumask_var (tmpmask);cpus_read_unlock ();
[email protected]
maintainerPATCH 158/168 powerpc/64s: Fix instruction encoding for lis in ppc_function_entry()
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Naveen N. Rao <[email protected]>
commit cea15316ceee2d4a51dfdecd79e08a438135416c upstream.
'lis r2,N' is 'addis r2,0,N' and the instruction encoding in the macro LIS_R2 is incorrect (it currently maps to 'addis r0,r2,N'). Fix the same.
--- arch/powerpc/include/asm/code-patching.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/include/asm/code-patching.h +1 -1
@@ -72,7 +72,7 @@ void __patch_exception(int exc, unsigned#endif #define OP_RT_RA_MASK 0xffff0000UL -#define LIS_R2 0x3c020000UL +#define LIS_R2 0x3c400000UL#define ADDIS_R2_R12 0x3c4c0000UL#define ADDI_R2_R2 0x38420000UL
[email protected]
maintainerPATCH 159/168 binfmt_misc: fix possible deadlock in bm_register_write
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Lior Ribak <[email protected]>
commit e7850f4d844e0acfac7e570af611d89deade3146 upstream.
There is a deadlock in bm_register_write:
First, in the begining of the function, a lock is taken on the binfmt_misc root inode with inode_lock(d_inode(root)).
Then, if the user used the MISC_FMT_OPEN_FILE flag, the function will call open_exec on the user-provided interpreter.
open_exec will call a path lookup, and if the path lookup process includes the root of binfmt_misc, it will try to take a shared lock on its inode again, but it is already locked, and the code will get stuck in a deadlock
To reproduce the bug:
$ echo ":iiiii:E::ii::/proc/sys/fs/binfmt_misc/bla:F" > /proc/sys/fs/binfmt_misc/register
backtrace of where the lock occurs (#5):
0 schedule () at ./arch/x86/include/asm/current.h:15 1 0xffffffff81b51237 in rwsem_down_read_slowpath (sem=0xffff888003b202e0, count=<optimized out>, state=state@entry=2) at kernel/locking/rwsem.c:992 2 0xffffffff81b5150a in __down_read_common (state=2, sem=<optimized out>) at kernel/locking/rwsem.c:1213 3 __down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1222 4 down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1355 5 0xffffffff811ee22a in inode_lock_shared (inode=<optimized out>) at ./include/linux/fs.h:783 6 open_last_lookups (op=0xffffc9000022fe34, file=0xffff888004098600, nd=0xffffc9000022fd10) at fs/namei.c:3177 7 path_openat (nd=nd@entry=0xffffc9000022fd10, op=op@entry=0xffffc9000022fe34, flags=flags@entry=65) at fs/namei.c:3366 8 0xffffffff811efe1c in do_filp_open (dfd=<optimized out>, pathname=pathname@entry=0xffff8880031b9000, op=op@entry=0xffffc9000022fe34) at fs/namei.c:3396 9 0xffffffff811e493f in do_open_execat (fd=fd@entry=-100, name=name@entry=0xffff8880031b9000, flags=<optimized out>, flags@entry=0) at fs/exec.c:913
10 0xffffffff811e4a92 in open_exec (name=<optimized out>) at fs/exec.c:948
11 0xffffffff8124aa84 in bm_register_write (file=<optimized out>, buffer=<optimized out>, count=19, ppos=<optimized out>) at fs/binfmt_misc.c:682 12 0xffffffff811decd2 in vfs_write (file=file@entry=0xffff888004098500, buf=buf@entry=0xa758d0 ":iiiii:E::ii::i:CF
", count=count@entry=19, pos=pos@entry=0xffffc9000022ff10) at fs/read_write.c:603
13 0xffffffff811defda in ksys_write (fd=<optimized out>, buf=0xa758d0 ":iiiii:E::ii::i:CF
", count=19) at fs/read_write.c:658
14 0xffffffff81b49813 in do_syscall_64 (nr=<optimized out>, regs=0xffffc9000022ff58) at arch/x86/entry/common.c:46
15 0xffffffff81c0007c in entry_SYSCALL_64 () at arch/x86/entry/entry_64.S:120
To solve the issue, the open_exec call is moved to before the write lock is taken by bm_register_write
--- fs/binfmt_misc.c | 29 ++++++++++++++--------------- 1 file changed, 14 insertions(+), 15 deletions(-)
fs/binfmt_misc.c +14 -15
@@ -696,12 +696,24 @@ static ssize_t bm_register_write(structstruct super_block *sb =file_inode (file)->i_sb;struct dentry *root = sb->s_root, *dentry;int err =0 ; +struct file *f =NULL ; e =create_entry (buffer, count);if (IS_ERR (e))return PTR_ERR (e); +if (e->flags &MISC_FMT_OPEN_FILE ) { + f =open_exec (e->interpreter); +if (IS_ERR (f)) { +pr_notice ("register: failed to install interpreter file %s\n" , + e->interpreter); +kfree (e); +return PTR_ERR (f); + } + e->interp_file = f; + } +inode_lock (d_inode (root)); dentry =lookup_one_len (e->name, root,strlen (e->name)); err =PTR_ERR (dentry);
@@ -725,21 +737,6 @@ static ssize_t bm_register_write(struct goto out2; } -if (e->flags &MISC_FMT_OPEN_FILE ) { -struct file *f; - - f =open_exec (e->interpreter); -if (IS_ERR (f)) { - err =PTR_ERR (f); -pr_notice ("register: failed to install interpreter file %s\n" , e->interpreter); -simple_release_fs (&bm_mnt, &entry_count); -iput (inode); - inode =NULL ; - goto out2; - } - e->interp_file = f; - } - e->dentry =dget (dentry); inode->i_private = e; inode->i_fop = &bm_entry_operations;
@@ -756,6 +753,8 @@ out:inode_unlock (d_inode (root));if (err) { +if (f) +filp_close (f,NULL );kfree (e);return err; }
[email protected]
maintainerPATCH 160/168 x86/unwind/orc: Disable KASAN checking in the ORC unwinder, part 2
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Josh Poimboeuf <[email protected]>
commit e504e74cc3a2c092b05577ce3e8e013fae7d94e6 upstream.
KASAN reserves "redzone" areas between stack frames in order to detect stack overruns. A read or write to such an area triggers a KASAN "stack-out-of-bounds" BUG.
Normally, the ORC unwinder stays in-bounds and doesn't access the redzone. But sometimes it can't find ORC metadata for a given instruction. This can happen for code which is missing ORC metadata, or for generated code. In such cases, the unwinder attempts to fall back to frame pointers, as a best-effort type thing.
This fallback often works, but when it doesn't, the unwinder can get confused and go off into the weeds into the KASAN redzone, triggering the aforementioned KASAN BUG.
But in this case, the unwinder's confusion is actually harmless and working as designed. It already has checks in place to prevent off-stack accesses, but those checks get short-circuited by the KASAN BUG. And a BUG is a lot more disruptive than a harmless unwinder warning.
Disable the KASAN checks by using READ_ONCE_NOCHECK() for all stack accesses. This finishes the job started by commit 881125bfe65b ("x86/unwind: Disable KASAN checking in the ORC unwinder"), which only partially fixed the issue.
--- arch/x86/kernel/unwind_orc.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
arch/x86/kernel/unwind_orc.c +6 -6
@@ -357,8 +357,8 @@ static bool deref_stack_regs(struct unwiif (!stack_access_ok (state, addr,sizeof (struct pt_regs )))return false; - *ip = regs->ip; - *sp = regs->sp; + *ip =READ_ONCE_NOCHECK (regs->ip); + *sp =READ_ONCE_NOCHECK (regs->sp);return true; }
@@ -370,8 +370,8 @@ static bool deref_stack_iret_regs(structif (!stack_access_ok (state, addr,IRET_FRAME_SIZE ))return false; - *ip = regs->ip; - *sp = regs->sp; + *ip =READ_ONCE_NOCHECK (regs->ip); + *sp =READ_ONCE_NOCHECK (regs->sp);return true; }
@@ -392,12 +392,12 @@ static bool get_reg(struct unwind_statereturn false;if (state->full_regs) { - *val = ((unsigned long *)state->regs)[reg]; + *val =READ_ONCE_NOCHECK (((unsigned long *)state->regs)[reg]);return true; }if (state->prev_regs) { - *val = ((unsigned long *)state->prev_regs)[reg]; + *val =READ_ONCE_NOCHECK (((unsigned long *)state->prev_regs)[reg]);return true; }
[email protected]
maintainerPATCH 161/168 KVM: arm64: Fix exclusive limit for IPA size
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Marc Zyngier <[email protected]>
commit 262b003d059c6671601a19057e9fe1a5e7f23722 upstream.
When registering a memslot, we check the size and location of that memslot against the IPA size to ensure that we can provide guest access to the whole of the memory.
Unfortunately, this check rejects memslot that end-up at the exact limit of the addressing capability for a given IPA size. For example, it refuses the creation of a 2GB memslot at 0x8000000 with a 32bit IPA space.
Fix it by relaxing the check to accept a memslot reaching the limit of the IPA space.
--- virt/kvm/arm/mmu.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
virt/kvm/arm/mmu.c +1 -2
@@ -2307,8 +2307,7 @@ int kvm_arch_prepare_memory_region(struc* Prevent userspace from creating a memory region outside of the IPA * space addressable by the KVM guest IPA space. */ -if (memslot->base_gfn + memslot->npages >= - (kvm_phys_size (kvm) >>PAGE_SHIFT )) +if ((memslot->base_gfn + memslot->npages) > (kvm_phys_size (kvm) >>PAGE_SHIFT ))return -EFAULT ;down_read (¤t->mm->mmap_sem);
[email protected]
maintainerPATCH 162/168 nvme: unlink head after removing last namespace
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Keith Busch <[email protected]>
commit d567572906d986dedb78b37f111c44eba033f3ef upstream.
The driver had been unlinking the namespace head from the subsystem's list only after the last reference was released, and outside of the list's subsys->lock protection.
There is no reason to track an empty head, so unlink the entry from the subsystem's list when the last namespace using that head is removed and with the mutex lock protecting the list update. The next namespace to attach reusing the previous NSID will allocate a new head rather than find the old head with mismatched identifiers.
--- drivers/nvme/host/core.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/nvme/host/core.c +5 -2
@@ -455,7 +455,6 @@ static void nvme_free_ns_head(struct krenvme_mpath_remove_disk (head);ida_simple_remove (&head->subsys->ns_ida, head->instance); -list_del_init (&head->entry);cleanup_srcu_struct (&head->srcu);nvme_put_subsystem (head->subsys);kfree (head);
@@ -3374,7 +3373,6 @@ static int __nvme_check_ids(struct nvme_list_for_each_entry (h, &subsys->nsheads, entry) {if (nvme_ns_ids_valid (&new->ids) && - !list_empty (&h->list) &&nvme_ns_ids_equal (&new->ids, &h->ids))return -EINVAL ; }
@@ -3629,6 +3627,8 @@ static int nvme_alloc_ns(struct nvme_ctr out_unlink_ns:mutex_lock (&ctrl->subsys->lock);list_del_rcu (&ns->siblings); +if (list_empty (&ns->head->list)) +list_del_init (&ns->head->entry);mutex_unlock (&ctrl->subsys->lock);nvme_put_ns_head (ns->head); out_free_id:
@@ -3651,7 +3651,10 @@ static void nvme_ns_remove(struct nvme_nmutex_lock (&ns->ctrl->subsys->lock);list_del_rcu (&ns->siblings); +if (list_empty (&ns->head->list)) +list_del_init (&ns->head->entry);mutex_unlock (&ns->ctrl->subsys->lock); +synchronize_rcu ();/* guarantee not available in head->list */ nvme_mpath_clear_current_path (ns);synchronize_srcu (&ns->head->srcu);/* wait for concurrent submissions */
[email protected]
maintainerPATCH 163/168 nvme: release namespace head reference on error
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Keith Busch <[email protected]>
commit ac262508daa88fb12c5dc53cf30bde163f9f26c9 upstream.
If a namespace identification does not match the subsystem's head for that NSID, release the reference that was taken when the matching head was initially found.
--- drivers/nvme/host/core.c | 1 + 1 file changed, 1 insertion(+)
drivers/nvme/host/core.c +1 -0
@@ -3467,6 +3467,7 @@ static int nvme_init_ns_head(struct nvme"IDs don't match for shared namespace %d\n" , nsid); ret = -EINVAL ; +nvme_put_ns_head (head); goto out_unlock; } }
[email protected]
maintainerPATCH 164/168 KVM: arm64: Ensure I-cache isolation between vcpus of a same VM
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Marc Zyngier <[email protected]>
Commit 01dc9262ff5797b675c32c0c6bc682777d23de05 upstream.
It recently became apparent that the ARMv8 architecture has interesting rules regarding attributes being used when fetching instructions if the MMU is off at Stage-1.
In this situation, the CPU is allowed to fetch from the PoC and allocate into the I-cache (unless the memory is mapped with the XN attribute at Stage-2).
If we transpose this to vcpus sharing a single physical CPU, it is possible for a vcpu running with its MMU off to influence another vcpu running with its MMU on, as the latter is expected to fetch from the PoU (and self-patching code doesn't flush below that level).
In order to solve this, reuse the vcpu-private TLB invalidation code to apply the same policy to the I-cache, nuking it every time the vcpu runs on a physical CPU that ran another vcpu of the same VM in the past.
This involve renaming __kvm_tlb_flush_local_vmid() to __kvm_flush_cpu_context(), and inserting a local i-cache invalidation there.
--- arch/arm/include/asm/kvm_asm.h | 2 +- arch/arm/kvm/hyp/tlb.c | 3 ++- arch/arm64/include/asm/kvm_asm.h | 2 +- arch/arm64/kvm/hyp/tlb.c | 3 ++- virt/kvm/arm/arm.c | 8 +++++++- 5 files changed, 13 insertions(+), 5 deletions(-)
arch/arm/include/asm/kvm_asm.h +1 -1
@@ -56,7 +56,7 @@ extern char __kvm_hyp_init_end[];extern void __kvm_flush_vm_context (void );extern void __kvm_tlb_flush_vmid_ipa (struct kvm *kvm,phys_addr_t ipa);extern void __kvm_tlb_flush_vmid (struct kvm *kvm); -extern void __kvm_tlb_flush_local_vmid (struct kvm_vcpu *vcpu); +extern void __kvm_flush_cpu_context (struct kvm_vcpu *vcpu);extern void __kvm_timer_set_cntvoff (u32 cntvoff_low,u32 cntvoff_high);
arch/arm/kvm/hyp/tlb.c +2 -1
@@ -45,7 +45,7 @@ void __hyp_text __kvm_tlb_flush_vmid_ipa__kvm_tlb_flush_vmid (kvm); } -void __hyp_text __kvm_tlb_flush_local_vmid(struct kvm_vcpu *vcpu) +void __hyp_text __kvm_flush_cpu_context(struct kvm_vcpu *vcpu) {struct kvm *kvm =kern_hyp_va (kern_hyp_va (vcpu)->kvm);
@@ -54,6 +54,7 @@ void __hyp_text __kvm_tlb_flush_local_vmisb ();write_sysreg (0 ,TLBIALL ); +write_sysreg (0 ,ICIALLU );dsb (nsh);isb ();
arch/arm64/include/asm/kvm_asm.h +1 -1
@@ -60,7 +60,7 @@ extern char __kvm_hyp_vector[];extern void __kvm_flush_vm_context (void );extern void __kvm_tlb_flush_vmid_ipa (struct kvm *kvm,phys_addr_t ipa);extern void __kvm_tlb_flush_vmid (struct kvm *kvm); -extern void __kvm_tlb_flush_local_vmid (struct kvm_vcpu *vcpu); +extern void __kvm_flush_cpu_context (struct kvm_vcpu *vcpu);extern void __kvm_timer_set_cntvoff (u32 cntvoff_low,u32 cntvoff_high);
arch/arm64/kvm/hyp/tlb.c +2 -1
@@ -182,7 +182,7 @@ void __hyp_text __kvm_tlb_flush_vmid(str__tlb_switch_to_host (kvm, &cxt); } -void __hyp_text __kvm_tlb_flush_local_vmid(struct kvm_vcpu *vcpu) +void __hyp_text __kvm_flush_cpu_context(struct kvm_vcpu *vcpu) {struct kvm *kvm =kern_hyp_va (kern_hyp_va (vcpu)->kvm);struct tlb_inv_context cxt;
@@ -191,6 +191,7 @@ void __hyp_text __kvm_tlb_flush_local_vm__tlb_switch_to_guest (kvm, &cxt);__tlbi (vmalle1); + asmvolatile ("ic iallu" );dsb (nsh);isb ();
virt/kvm/arm/arm.c +7 -1
@@ -373,11 +373,17 @@ void kvm_arch_vcpu_load(struct kvm_vcpu cpu_data =this_cpu_ptr (&kvm_host_data);/* +* We guarantee that both TLBs and I-cache are private to each +* vcpu. If detecting that a vcpu from the same VM has +* previously run on the same physical CPU, call into the +* hypervisor code to nuke the relevant contexts. +* +* We might get preempted before the vCPU actually runs, but * We might get preempted before the vCPU actually runs, but * over-invalidation doesn't affect correctness. */ if (*last_ran != vcpu->vcpu_id) { -kvm_call_hyp (__kvm_tlb_flush_local_vmid, vcpu); +kvm_call_hyp (__kvm_flush_cpu_context, vcpu); *last_ran = vcpu->vcpu_id; }
[email protected]
maintainerPATCH 165/168 KVM: arm64: Reject VM creation when the default IPA size is unsupported
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Marc Zyngier <[email protected]>
Commit 7d717558dd5ef10d28866750d5c24ff892ea3778 upstream.
KVM/arm64 has forever used a 40bit default IPA space, partially due to its 32bit heritage (where the only choice is 40bit).
However, there are implementations in the wild that have a *cough* much smaller *cough* IPA space, which leads to a misprogramming of VTCR_EL2, and a guest that is stuck on its first memory access if userspace dares to ask for the default IPA setting (which most VMMs do).
Instead, blundly reject the creation of such VM, as we can't satisfy the requirements from userspace (with a one-off warning). Also clarify the boot warning, and document that the VM creation will fail when an unsupported IPA size is provided.
Although this is an ABI change, it doesn't really change much for userspace:
- the guest couldn't run before this change, but no error was returned. At least userspace knows what is happening.
- a memory slot that was accepted because it did fit the default IPA space now doesn't even get a chance to be registered.
The other thing that is left doing is to convince userspace to actually use the IPA space setting instead of relying on the antiquated default.
--- Documentation/virt/kvm/api.txt | 3 +++ arch/arm64/kvm/reset.c | 11 ++++++++--- 2 files changed, 11 insertions(+), 3 deletions(-)
Documentation/virt/kvm/api.txt +3 -0
@@ -172,6 +172,9 @@ is dependent on the CPU capability and t be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION ioctl() at run-time. +Creation of the VM will fail if the requested IPA size (whether it is +implicit or explicit) is unsupported on the host. + Please note that configuring the IPA size does not affect the capability exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects size of the address translated by the stage2 level (guest physical to
arch/arm64/kvm/reset.c +8 -3
@@ -378,10 +378,10 @@ void kvm_set_ipa_limit(void)pr_info ("kvm: Limiting the IPA size due to kernel %s Address limit\n" , (va_max < pa_max) ?"Virtual" :"Physical" ); -WARN (ipa_max <KVM_PHYS_SHIFT , -"KVM IPA limit (%d bit) is smaller than default size\n" , ipa_max); kvm_ipa_limit = ipa_max; -kvm_info ("IPA Size Limit: %dbits\n" , kvm_ipa_limit); +kvm_info ("IPA Size Limit: %d bits%s\n" , kvm_ipa_limit, + ((kvm_ipa_limit <KVM_PHYS_SHIFT ) ? +" (Reduced IPA size, limited VM/VMM compatibility)" :"" )); }/*
@@ -408,6 +408,11 @@ int kvm_arm_setup_stage2(struct kvm *kvmreturn -EINVAL ; }else { phys_shift =KVM_PHYS_SHIFT ; +if (phys_shift > kvm_ipa_limit) { +pr_warn_once ("%s using unsupported default IPA limit, upgrade your VMM\n" , + current->comm); +return -EINVAL ; + } } parange =read_sanitised_ftr_reg (SYS_ID_AA64MMFR0_EL1 ) &7 ;
[email protected]
maintainerPATCH 166/168 xen/events: reset affinity of 2-level event when tearing it down
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Juergen Gross <[email protected]>
commit 9e77d96b8e2724ed00380189f7b0ded61113b39f upstream.
When creating a new event channel with 2-level events the affinity needs to be reset initially in order to avoid using an old affinity from earlier usage of the event channel port. So when tearing an event channel down reset all affinity bits.
The same applies to the affinity when onlining a vcpu: all old affinity settings for this vcpu must be reset. As percpu events get initialized before the percpu event channel hook is called, resetting of the affinities happens after offlining a vcpu (this is working, as initial percpu memory is zeroed out).
--- drivers/xen/events/events_2l.c | 15 +++++++++++++++ drivers/xen/events/events_base.c | 1 + drivers/xen/events/events_internal.h | 8 ++++++++ 3 files changed, 24 insertions(+)
drivers/xen/events/events_2l.c +15 -0
@@ -47,6 +47,11 @@ static unsigned evtchn_2l_max_channels(vreturn EVTCHN_2L_NR_CHANNELS ; } +static void evtchn_2l_remove (evtchn_port_t evtchn,unsigned int cpu) +{ +clear_bit (evtchn,BM (per_cpu (cpu_evtchn_mask, cpu))); +} +static void evtchn_2l_bind_to_cpu (struct irq_info *info,unsigned cpu) {clear_bit (info->evtchn,BM (per_cpu (cpu_evtchn_mask, info->cpu)));
@@ -354,9 +359,18 @@ static void evtchn_2l_resume(void) EVTCHN_2L_NR_CHANNELS/BITS_PER_EVTCHN_WORD); } +static int evtchn_2l_percpu_deinit (unsigned int cpu) +{ +memset (per_cpu (cpu_evtchn_mask, cpu),0 ,sizeof (xen_ulong_t) * +EVTCHN_2L_NR_CHANNELS /BITS_PER_EVTCHN_WORD ); + +return 0 ; +} +static const struct evtchn_ops evtchn_ops_2l = { .max_channels = evtchn_2l_max_channels, .nr_channels = evtchn_2l_max_channels, + .remove = evtchn_2l_remove, .bind_to_cpu = evtchn_2l_bind_to_cpu, .clear_pending = evtchn_2l_clear_pending, .set_pending = evtchn_2l_set_pending,
@@ -366,6 +380,7 @@ static const struct evtchn_ops evtchn_op .unmask = evtchn_2l_unmask, .handle_events = evtchn_2l_handle_events, .resume = evtchn_2l_resume, + .percpu_deinit = evtchn_2l_percpu_deinit, };void __init xen_evtchn_2l_init (void )
drivers/xen/events/events_base.c +1 -0
@@ -286,6 +286,7 @@ static int xen_irq_info_pirq_setup(unsigstatic void xen_irq_info_cleanup (struct irq_info *info) {set_evtchn_to_irq (info->evtchn,-1 ); +xen_evtchn_port_remove (info->evtchn, info->cpu); info->evtchn =0 ; }
drivers/xen/events/events_internal.h +8 -0
@@ -65,6 +65,7 @@ struct evtchn_ops {unsigned (*nr_channels)(void );int (*setup)(struct irq_info *info); +void (*remove)(evtchn_port_t port,unsigned int cpu);void (*bind_to_cpu)(struct irq_info *info,unsigned cpu);void (*clear_pending)(unsigned port);
@@ -107,6 +108,13 @@ static inline int xen_evtchn_port_setup(return 0 ; } +static inline void xen_evtchn_port_remove (evtchn_port_t evtchn, +unsigned int cpu) +{ +if (evtchn_ops->remove) + evtchn_ops->remove (evtchn, cpu); +} +static inline void xen_evtchn_port_bind_to_cpu (struct irq_info *info,unsigned cpu) {
[email protected]
maintainerPATCH 167/168 xen/events: dont unmask an event channel when an eoi is pending
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Juergen Gross <[email protected]>
commit 25da4618af240fbec6112401498301a6f2bc9702 upstream.
An event channel should be kept masked when an eoi is pending for it. When being migrated to another cpu it might be unmasked, though.
In order to avoid this keep three different flags for each event channel to be able to distinguish "normal" masking/unmasking from eoi related masking/unmasking and temporary masking. The event channel should only be able to generate an interrupt if all flags are cleared.
[boris -- corrected Fixed tag format]
--- drivers/xen/events/events_2l.c | 7 -- drivers/xen/events/events_base.c | 110 ++++++++++++++++++++++++++--------- drivers/xen/events/events_fifo.c | 7 -- drivers/xen/events/events_internal.h | 13 +--- 4 files changed, 88 insertions(+), 49 deletions(-)
drivers/xen/events/events_2l.c +0 -7
@@ -76,12 +76,6 @@ static bool evtchn_2l_is_pending(unsignereturn sync_test_bit (port,BM (&s->evtchn_pending[0 ])); } -static bool evtchn_2l_test_and_set_mask (unsigned port) -{ -struct shared_info *s = HYPERVISOR_shared_info; -return sync_test_and_set_bit (port,BM (&s->evtchn_mask[0 ])); -} -static void evtchn_2l_mask (unsigned port) {struct shared_info *s = HYPERVISOR_shared_info;
@@ -375,7 +369,6 @@ static const struct evtchn_ops evtchn_op .clear_pending = evtchn_2l_clear_pending, .set_pending = evtchn_2l_set_pending, .is_pending = evtchn_2l_is_pending, - .test_and_set_mask = evtchn_2l_test_and_set_mask, .mask = evtchn_2l_mask, .unmask = evtchn_2l_unmask, .handle_events = evtchn_2l_handle_events,
drivers/xen/events/events_base.c +82 -28
237 lines not on this page · raw message
drivers/xen/events/events_fifo.c +0 -7
@@ -209,12 +209,6 @@ static bool evtchn_fifo_is_pending(unsigreturn sync_test_bit (EVTCHN_FIFO_BIT (PENDING , word),BM (word)); } -static bool evtchn_fifo_test_and_set_mask (unsigned port) -{ -event_word_t *word =event_word_from_port (port); -return sync_test_and_set_bit (EVTCHN_FIFO_BIT (MASKED , word),BM (word)); -} -static void evtchn_fifo_mask (unsigned port) {event_word_t *word =event_word_from_port (port);
@@ -420,7 +414,6 @@ static const struct evtchn_ops evtchn_op .clear_pending = evtchn_fifo_clear_pending, .set_pending = evtchn_fifo_set_pending, .is_pending = evtchn_fifo_is_pending, - .test_and_set_mask = evtchn_fifo_test_and_set_mask, .mask = evtchn_fifo_mask, .unmask = evtchn_fifo_unmask, .handle_events = evtchn_fifo_handle_events,
drivers/xen/events/events_internal.h +6 -7
@@ -33,13 +33,18 @@ struct irq_info {struct list_head eoi_list;short refcnt;short spurious_cnt; -enum xen_irq_type type;/* type */ +short type;/* type */ +u8 mask_reason;/* Why is event channel masked */ +#define EVT_MASK_REASON_EXPLICIT 0x01 +#define EVT_MASK_REASON_TEMPORARY 0x02 +#define EVT_MASK_REASON_EOI_PENDING 0x04unsigned irq;unsigned int evtchn;/* event channel */ unsigned short cpu;/* cpu bound */ unsigned short eoi_cpu;/* EOI must happen on this cpu */ unsigned int irq_epoch;/* If eoi_cpu valid: irq_epoch of event */ u64 eoi_time;/* Time in jiffies when to EOI. */ +spinlock_t lock;union {unsigned short virq;
@@ -71,7 +76,6 @@ struct evtchn_ops {void (*clear_pending)(unsigned port);void (*set_pending)(unsigned port);bool (*is_pending)(unsigned port); -bool (*test_and_set_mask)(unsigned port);void (*mask)(unsigned port);void (*unmask)(unsigned port);
@@ -136,11 +140,6 @@ static inline bool test_evtchn(unsignedreturn evtchn_ops->is_pending (port); } -static inline bool test_and_set_mask (unsigned port) -{ -return evtchn_ops->test_and_set_mask (port); -} -static inline void mask_evtchn (unsigned port) {return evtchn_ops->mask (port);
[email protected]
maintainerPATCH 168/168 xen/events: avoid handling the same event on two cpus at the same time
In reply to [email protected]From: Greg Kroah-Hartman <[email protected]>
From: Juergen Gross <[email protected]>
commit b6622798bc50b625a1e62f82c7190df40c1f5b21 upstream.
When changing the cpu affinity of an event it can happen today that (with some unlucky timing) the same event will be handled on the old and the new cpu at the same time.
Avoid that by adding an "event active" flag to the per-event data and call the handler only if this flag isn't set.
--- drivers/xen/events/events_base.c | 25 +++++++++++++++++-------- drivers/xen/events/events_internal.h | 1 + 2 files changed, 18 insertions(+), 8 deletions(-)
drivers/xen/events/events_base.c +17 -8
@@ -694,6 +694,12 @@ static void xen_evtchn_close(unsigned inBUG (); } +static void event_handler_exit (struct irq_info *info) +{ +smp_store_release (&info->is_active,0 ); +clear_evtchn (info->evtchn); +} +static void pirq_query_unmask (int irq) {struct physdev_irq_status_query irq_status;
@@ -724,13 +730,13 @@ static void eoi_pirq(struct irq_data *da likely(!irqd_irq_disabled(data))) {do_mask (info,EVT_MASK_REASON_TEMPORARY ); -clear_evtchn (evtchn); +event_handler_exit (info);irq_move_masked_irq (data);do_unmask (info,EVT_MASK_REASON_TEMPORARY ); } else -clear_evtchn (evtchn ); +event_handler_exit (info );if (pirq_needs_eoi (data->irq)) { rc =HYPERVISOR_physdev_op (PHYSDEVOP_eoi, &eoi);
@@ -1566,6 +1572,8 @@ void handle_irq_for_port(evtchn_port_t p } info =info_for_irq (irq); +if (xchg_acquire (&info->is_active,1 )) +return ;if (ctrl->defer_eoi) { info->eoi_cpu =smp_processor_id ();
@@ -1753,13 +1761,13 @@ static void ack_dynirq(struct irq_data * likely(!irqd_irq_disabled(data))) {do_mask (info,EVT_MASK_REASON_TEMPORARY ); -clear_evtchn (evtchn); +event_handler_exit (info);irq_move_masked_irq (data);do_unmask (info,EVT_MASK_REASON_TEMPORARY ); } else -clear_evtchn (evtchn ); +event_handler_exit (info ); }static void mask_ack_dynirq (struct irq_data *data)
@@ -1775,7 +1783,7 @@ static void lateeoi_ack_dynirq(struct irif (VALID_EVTCHN (evtchn)) {do_mask (info,EVT_MASK_REASON_EOI_PENDING ); -clear_evtchn (evtchn); +event_handler_exit (info); } }
@@ -1786,7 +1794,7 @@ static void lateeoi_mask_ack_dynirq(struif (VALID_EVTCHN (evtchn)) {do_mask (info,EVT_MASK_REASON_EXPLICIT ); -clear_evtchn (evtchn); +event_handler_exit (info); } }
@@ -1895,10 +1903,11 @@ static void restore_cpu_ipis(unsigned in/* Clear an irq's pending state, in preparation for polling on it */ void xen_clear_irq_pending (int irq) { -int evtchn =evtchn_from_irq (irq); +struct irq_info *info =info_for_irq (irq); +evtchn_port_t evtchn = info ? info->evtchn :0 ;if (VALID_EVTCHN (evtchn)) -clear_evtchn (evtchn); +event_handler_exit (info); }EXPORT_SYMBOL (xen_clear_irq_pending);void xen_set_irq_pending (int irq)
drivers/xen/events/events_internal.h +1 -0
@@ -38,6 +38,7 @@ struct irq_info {#define EVT_MASK_REASON_EXPLICIT 0x01#define EVT_MASK_REASON_TEMPORARY 0x02#define EVT_MASK_REASON_EOI_PENDING 0x04 +u8 is_active;/* Is event just being handled? */ unsigned irq;unsigned int evtchn;/* event channel */ unsigned short cpu;/* cpu bound */
Quoted 11 lines · On Mon, 15 Mar 2021 14:53:52 +0100, [email protected] wrote:
From: Greg Kroah-Hartman <[email protected]>
This is the start of the stable review cycle for the 5.4.106 release. There are 168 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 Wed, 17 Mar 2021 13:55:26 +0000. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
Quoted 8 lines
https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.4.106-rc1.gzor in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.gitlinux-5.4.y
and the diffstat can be found below.thanks,
greg k-h
All tests passing for Tegra ...
Test results for stable-v5.4:
12 builds: 12 pass, 0 fail
26 boots: 26 pass, 0 fail
57 tests: 57 pass, 0 failLinux version: 5.4.106-rc1-gfed4321947f0
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,
tegra194-p2972-0000, tegra20-ventana,
tegra210-p2371-2180, tegra210-p3450-0000,
tegra30-cardhu-a04Jon
Quoted 11 lines · On 3/15/2021 6:53 AM, [email protected] wrote:
From: Greg Kroah-Hartman <[email protected]>
This is the start of the stable review cycle for the 5.4.106 release. There are 168 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 Wed, 17 Mar 2021 13:55:26 +0000. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
Quoted 8 lines
https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.4.106-rc1.gzor in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.gitlinux-5.4.y
and the diffstat can be found below.thanks,
greg k-h
On ARCH_BRCMSTB, using 32-bit ARM and 64-bit kernels:
Signature · 1 line
Florian
Quoted 3 lines · On Mon, 15 Mar 2021 14:53:52 +0100 [email protected] wrote:
From: Greg Kroah-Hartman <[email protected]>
This is the start of the stable review cycle for the 5.4.106 release.
Tested on amd64, arm64, armhf, i386, m68k, or1k, powerpc, ppc64, ppc64el, riscv64, s390x, sparc64. No problems detected.