4.18.8-stable review
Subsystems (4)
Greg Kroah-Hartman
maintainercover 4.18.8-stable review
This is the start of the stable review cycle for the 4.18.8 release. There are 197 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 Sat Sep 15 13:17:57 UTC 2018. 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/v4.x/stable-review/patch-4.18.8-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-4.18.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 4.18.8-rc1
Gustavo A. R. Silva <[email protected]>
ASoC: wm8994: Fix missing break in switch
Robert Munteanu <[email protected]>
HID: redragon: fix num lock and caps lock LEDs
Arnd Bergmann <[email protected]>
x86: kvm: avoid unused variable warning
Junaid Shahid <[email protected]>
kvm: x86: Set highest physical address bits in non-present/reserved SPTEs
Randy Dunlap <[email protected]>
kbuild: make missing $DEPMOD a Warning instead of an Error
Juergen Gross <[email protected]>
x86/xen: don't write ptes directly in 32-bit PV guests
Juergen Gross <[email protected]>
x86/pae: use 64 bit atomic xchg function in native_ptep_get_and_clear
Joel Fernandes (Google) <[email protected]>
debugobjects: Make stack check warning more informative
Michel Dänzer <[email protected]>
drm/amdgpu: Don't warn on destroying a pinned BO
Michel Dänzer <[email protected]>
drm/amdgpu: Warn and update pin_size values when destroying a pinned BO
Michel Dänzer <[email protected]>
drm/amdgpu: Make pin_size values atomic
Michel Dänzer <[email protected]>
drm/amdgpu: Keep track of amount of pinned CPU visible VRAM
Chuanhua Lei <[email protected]>
x86/tsc: Prevent result truncation on 32bit
Jani Nikula <[email protected]>
drm/i915: set DP Main Stream Attribute for color range on DDI platforms
Nadav Amit <[email protected]>
mm: respect arch_dup_mmap() return value
Randy Dunlap <[email protected]>
uapi/linux/keyctl.h: don't use C++ reserved keyword as a struct member name
Jan-Marek Glogowski <[email protected]>
drm/i915: Re-apply "Perform link quality check, unconditionally during long pulse"
Christian König <[email protected]>
drm/amdgpu: fix incorrect use of drm_file->pid
Christian König <[email protected]>
drm/amdgpu: fix incorrect use of fcheck
Likun Gao <[email protected]>
drm/amdgpu:add VCN booting with firmware loaded by PSP
Likun Gao <[email protected]>
drm/amdgpu:add VCN support in PSP driver
Likun Gao <[email protected]>
drm/amdgpu:add new firmware id for VCN
James Zhu <[email protected]>
drm/amdgpu:add tmr mc address into amdgpu_firmware_info
James Zhu <[email protected]>
drm/amdgpu: update tmr mc address
Mikita Lipski <[email protected]>
drm/amd/display: Check if clock source in use before disabling
Mikita Lipski <[email protected]>
drm/amd/display: Pass connector id when executing VBIOS CT
Sandy Huang <[email protected]>
drm/rockchip: vop: fix irq disabled after vop driver probed
Heiko Stuebner <[email protected]>
drm/rockchip: vop: split out core clock enablement into separate functions
Julia Lawall <[email protected]>
drm/rockchip: lvds: add missing of_node_put
Harry Wentland <[email protected]>
drm/amd/display: Report non-DP display as disconnected without EDID
Leo (Sunpeng) Li <[email protected]>
drm/amd/display: Use requested HDMI aspect ratio
Mikita Lipski <[email protected]>
drm/amd/display: update clk for various HDMI color depths
Mikita Lipski <[email protected]>
drm/amd/display: Don't share clk source between DP and HDMI
Gustavo A. R. Silva <[email protected]>
drm/amd/display: fix type of variable
Kai-Heng Feng <[email protected]>
drm/edid: Add 6 bpc quirk for SDC panel in Lenovo B50-80
Lubosz Sarnecki <[email protected]>
drm/edid: Quirk Vive Pro VR headset non-desktop.
Rex Zhu <[email protected]>
drm/amd/pp/Polaris12: Fix a chunk of registers missed to program
Evan Quan <[email protected]>
drm/amd/powerplay: fixed uninitialized value
Rex Zhu <[email protected]>
drm/amd/pp: Convert voltage unit in mV*4 to mV on CZ/ST
Michel Dänzer <[email protected]>
drm/amdgpu: Fix RLC safe mode test in gfx_v9_0_enter_rlc_safe_mode
Rex Zhu <[email protected]>
drm/amdgpu: fix a reversed condition
Alex Deucher <[email protected]>
drm/amdgpu: update uvd_v6_0_ring_vm_funcs to use new nop packet
Rodrigo Vivi <[email protected]>
drm/i915: Free write_buf that we allocated with kzalloc.
Fredrik Schön <[email protected]>
drm/i915: Increase LSPCON timeout
Ville Syrjälä <[email protected]>
drm/i915: Nuke the LVDS lid notifier
Chris Wilson <[email protected]>
drm/i915/lpe: Mark LPE audio runtime pm as "no callbacks"
David Sterba <[email protected]>
btrfs: fix mount and ioctl device scan ioctl race
David Sterba <[email protected]>
btrfs: reorder initialization before the mount locks uuid_mutex
David Sterba <[email protected]>
btrfs: lift uuid_mutex to callers of btrfs_parse_early_options
David Sterba <[email protected]>
btrfs: lift uuid_mutex to callers of btrfs_scan_one_device
Anand Jain <[email protected]>
btrfs: use device_list_mutex when removing stale devices
Anand Jain <[email protected]>
btrfs: rename local devices for fs_devices in btrfs_free_stale_devices(
Anand Jain <[email protected]>
btrfs: extend locked section when adding a new device in device_list_add
Anand Jain <[email protected]>
btrfs: do btrfs_free_stale_devices outside of device_list_add
Marc Zyngier <[email protected]>
ARM: rockchip: Force CONFIG_PM on Rockchip systems
Marc Zyngier <[email protected]>
arm64: rockchip: Force CONFIG_PM on Rockchip systems
Bart Van Assche <[email protected]>
btrfs: Fix a C compliance issue
Qu Wenruo <[email protected]>
btrfs: Don't remove block group that still has pinned down bytes
David Sterba <[email protected]>
btrfs: lift uuid_mutex to callers of btrfs_open_devices
Qu Wenruo <[email protected]>
btrfs: check-integrity: Fix NULL pointer dereference for degraded mount
Qu Wenruo <[email protected]>
btrfs: tree-checker: Detect invalid and empty essential trees
Qu Wenruo <[email protected]>
btrfs: relocation: Only remove reloc rb_trees if reloc control has been initialized
Anand Jain <[email protected]>
btrfs: fix in-memory value of total_devices after seed device deletion
Misono Tomohiro <[email protected]>
btrfs: replace: Reset on-disk dev stats value after replace
Qu Wenruo <[email protected]>
btrfs: Exit gracefully when chunk map cannot be inserted to the tree
Lucas Stach <[email protected]>
drm/etnaviv: fix crash in GPU suspend when init failed due to buffer placement
Jim Mattson <[email protected]>
kvm: nVMX: Fix fault vector for VMX operation at CPL > 0
Sean Christopherson <[email protected]>
KVM: vmx: track host_state.loaded using a loaded_vmcs pointer
David Francis <[email protected]>
drm/amd/display: Read back max backlight value at boot
Levin Du <[email protected]>
clk: rockchip: Add pclk_rkpwm_pmu to PMU critical clocks in rk3399
Mahesh Salgaonkar <[email protected]>
powerpc/pseries: Avoid using the size greater than RTAS_ERROR_LOG_MAX.
Masahiro Yamada <[email protected]>
um: fix parallel building with O= option
Michael Ellerman <[email protected]>
powerpc/64s: Make rfi_flush_fallback a little more robust
Randy Dunlap <[email protected]>
powerpc/platforms/85xx: fix t1042rdb_diu.c build errors & warning
Steve French <[email protected]>
smb3: if server does not support posix do not allow posix mount option
Steve French <[email protected]>
SMB3: Number of requests sent should be displayed for SMB3 not just CIFS
Aurelien Aptel <[email protected]>
CIFS: fix memory leak and remove dead code
Steve French <[email protected]>
smb3: fix reset of bytes read and written stats
Bart Van Assche <[email protected]>
cfq: Suppress compiler warnings about comparisons
YueHaibing <[email protected]>
RDS: IB: fix 'passing zero to ERR_PTR()' warning
nixiaoming <[email protected]>
selinux: cleanup dentry and inodes on error in selinuxfs
Breno Leitao <[email protected]>
selftests/powerpc: Kill child processes on SIGINT
Ralf Goebel <[email protected]>
iommu/omap: Fix cache flushes on L2 table entries
Matthias Kaehlcke <[email protected]>
ASoC: rt5677: Fix initialization of rt5677_of_match.data
Ian Abbott <[email protected]>
staging: comedi: ni_mio_common: fix subdevice flags for PFI subdevice
John Pittman <[email protected]>
dm kcopyd: avoid softlockup in run_complete_job
Thomas Petazzoni <[email protected]>
PCI: mvebu: Fix I/O space end address calculation
Roger Pau Monne <[email protected]>
xen/balloon: fix balloon initialization for PVH Dom0
Dmitry Torokhov <[email protected]>
Input: do not use WARN() in input_alloc_absinfo()
Wei Yongjun <[email protected]>
NFSv4: Fix error handling in nfs4_sp4_select_mode()
Dan Carpenter <[email protected]>
scsi: aic94xx: fix an error code in aic94xx_init()
Jianchao Wang <[email protected]>
blk-mq: count the hctx as active before allocating tag
Hans de Goede <[email protected]>
ACPI / scan: Initialize status to ACPI_STA_DEFAULT
Stefan Haberland <[email protected]>
s390/dasd: fix panic for failed online processing
Stefan Haberland <[email protected]>
s390/dasd: fix hanging offline processing due to canceled worker
Winnie Chang <[email protected]>
brcmfmac: fix brcmf_wiphy_wowl_params() NULL pointer dereference
Greg Edwards <[email protected]>
block: bvec_nr_vecs() returns value for wrong slab
Sandipan Das <[email protected]>
perf probe powerpc: Fix trace event post-processing
Dan Carpenter <[email protected]>
powerpc: Fix size calculation using resource_size()
Michael Ellerman <[email protected]>
powerpc/uaccess: Enable get_user(u64, *p) on 32-bit
Yonghong Song <[email protected]>
bpf: fix bpffs non-array map seq_show issue
Anton Vasilyev <[email protected]>
pinctrl: axp209: Fix NULL pointer dereference after allocation
Chao Yu <[email protected]>
f2fs: fix to clear PG_checked flag in set_page_dirty()
Jean-Philippe Brucker <[email protected]>
net/9p: fix error path of p9_virtio_probe
Tomas Bortoli <[email protected]>
net/9p/trans_fd.c: fix race by holding the lock
Jonas Gorski <[email protected]>
irqchip/bcm7038-l1: Hide cpu offline callback when building for !SMP
Dan Carpenter <[email protected]>
irqchip/stm32: Fix init error handling
Palmer Dabbelt <[email protected]>
RISC-V: Use KBUILD_CFLAGS instead of KCFLAGS when building the vDSO
Chao Yu <[email protected]>
f2fs: fix avoid race between truncate and background GC
Chao Yu <[email protected]>
f2fs: avoid race between zero_range and background GC
Benno Evers <[email protected]>
perf tools: Check for null when copying nsinfo.
Denis Efremov <[email protected]>
coccicheck: return proper error code on fail
Nicholas Kazlauskas <[email protected]>
drm/amd/display: Guard against null crtc in CRC IRQ
Myron Stowe <[email protected]>
PCI: Match Root Port's MPS to endpoint's MPSS as necessary
Jian Shen <[email protected]>
net: hns3: Fix for phy link issue when using marvell phy driver
Jens Axboe <[email protected]>
block: don't warn for flush on read-only device
Xi Wang <[email protected]>
net: hns3: Fix for command format parsing error in hclge_is_all_function_id_zero
Kim Phillips <[email protected]>
perf arm spe: Fix uninitialized record error variable
Erik Schmauss <[email protected]>
ACPICA: ACPICA: add status check for acpi_hw_read before assigning return value
Gal Pressman <[email protected]>
RDMA/hns: Fix usage of bitmap allocation functions return values
Richard Weinberger <[email protected]>
ubi: Initialize Fastmap checkmapping correctly
Daniel Borkmann <[email protected]>
tcp, ulp: add alias for all ulp modules
Florian Westphal <[email protected]>
netfilter: fix memory leaks on netlink_dump_start error
Aleh Filipovich <[email protected]>
platform/x86: asus-nb-wmi: Add keymap entry for lid flip action on UX360
Michal Hocko <[email protected]>
netfilter: x_tables: do not fail xt_alloc_table_info too easilly
Guenter Roeck <[email protected]>
mfd: sm501: Set coherent_dma_mask when creating subdevices
Tan Hu <[email protected]>
ipvs: fix race between ip_vs_conn_new() and ip_vs_del_dest()
Philipp Rudo <[email protected]>
s390/kdump: Fix memleak in nt_vmcoreinfo
Florian Westphal <[email protected]>
netfilter: ip6t_rpfilter: set F_IFACE for linklocal addresses
Jesper Dangaard Brouer <[email protected]>
samples/bpf: all XDP samples should unload xdp/bpf prog on SIGTERM
Daniel Borkmann <[email protected]>
bpf, sockmap: fix leakage of smap_psock_map_entry
Tariq Toukan <[email protected]>
net/xdp: Fix suspicious RCU usage warning
Daniel Borkmann <[email protected]>
bpf, sockmap: fix sock_map_ctx_update_elem race with exist/noexist
Daniel Borkmann <[email protected]>
tcp, ulp: fix leftover icsk_ulp_ops preventing sock from reattach
Daniel Borkmann <[email protected]>
bpf, sockmap: fix map elem deletion race with smap_stop_sock
Randy Dunlap <[email protected]>
platform/x86: intel_punit_ipc: fix build errors
Tetsuo Handa <[email protected]>
fs/dcache.c: fix kmemcheck splat at take_dentry_name_snapshot()
Mike Rapoport <[email protected]>
mm: make DEFERRED_STRUCT_PAGE_INIT explicitly depend on SPARSEMEM
Andrey Ryabinin <[email protected]>
mm/fadvise.c: fix signed overflow UBSAN complaint
Srikar Dronamraju <[email protected]>
powerpc/topology: Get topology for shared processors at boot
Jerome Brunet <[email protected]>
pwm: meson: Fix mux clock names
Michael J. Ruhl <[email protected]>
IB/hfi1: Invalid NUMA node information can cause a divide by zero
Hans de Goede <[email protected]>
i2c: core: ACPI: Make acpi_gsb_i2c_read_bytes() check i2c_transfer return value
Arnd Bergmann <[email protected]>
x86/mce: Add notifier_block forward declaration
Suzuki K Poulose <[email protected]>
virtio: pci-legacy: Validate queue pfn
Dan Carpenter <[email protected]>
apparmor: fix an error code in __aa_create_ns()
Randy Dunlap <[email protected]>
scripts: modpost: check memory allocation results
Johannes Berg <[email protected]>
workqueue: re-add lockdep dependencies for flushing
Johannes Berg <[email protected]>
workqueue: skip lockdep wq dependency in cancel_work_sync()
OGAWA Hirofumi <[email protected]>
fat: validate ->i_start before using
James Morse <[email protected]>
fs/proc/kcore.c: use __pa_symbol() for KCORE_TEXT list entries
Marc Zyngier <[email protected]>
iommu/rockchip: Move irq request past pm_runtime_enable
Ernesto A. Fernández <[email protected]>
hfsplus: fix NULL dereference in hfsplus_lookup()
Marc Zyngier <[email protected]>
iommu/rockchip: Handle errors returned from PM framework
Arnd Bergmann <[email protected]>
reiserfs: change j_timestamp type to time64_t
Arnd Bergmann <[email protected]>
fs/proc/vmcore.c: hide vmcoredd_mmap_dumps() for nommu builds
Jann Horn <[email protected]>
fork: don't copy inconsistent signal handler state to child
Laura Abbott <[email protected]>
sunrpc: Don't use stack buffer with scatterlist
Ernesto A. Fernández <[email protected]>
hfs: prevent crash on exit from failed search
Tetsuo Handa <[email protected]>
hfsplus: don't return 0 when fill_super() failed
Ronnie Sahlberg <[email protected]>
cifs: check if SMB2 PDU size has been padded and suppress the warning
Stephen Hemminger <[email protected]>
hv_netvsc: ignore devices that are not PCI
Jason Wang <[email protected]>
vhost: correctly check the iova range when waking virtqueue
Ido Schimmel <[email protected]>
mlxsw: spectrum_switchdev: Do not leak RIFs when removing bridge
Xin Long <[email protected]>
sctp: hold transport before accessing its asoc in sctp_transport_get_next
Jakub Kicinski <[email protected]>
nfp: wait for posted reconfigs when disabling the device
Haishuang Yan <[email protected]>
ip6_vti: fix a null pointer deference when destroy vti6 tunnel
Haishuang Yan <[email protected]>
ip6_vti: fix creating fallback tunnel device for vti6
Jerome Brunet <[email protected]>
Revert "net: stmmac: Do not keep rearming the coalesce timer in stmmac_xmit"
Azat Khuzhin <[email protected]>
r8169: set RxConfig after tx/rx is enabled for RTL8169sb/8110sb devices
Tariq Toukan <[email protected]>
net/mlx5: Fix SQ offset in QPs with small RQ
David Ahern <[email protected]>
net/ipv6: Put lwtstate when destroying fib6_info
David Ahern <[email protected]>
net/ipv6: Only update MTU metric if it set
Hangbin Liu <[email protected]>
net/ipv6: init ip6 anycast rt->dst.input as ip6_input
Alexey Kodanev <[email protected]>
ipv6: don't get lwtstate twice in ip6_rt_copy_init()
Ahmad Fatoum <[email protected]>
net: macb: Fix regression breaking non-MDIO fixed-link PHYs
Xin Long <[email protected]>
erspan: set erspan_ver to 1 by default when adding an erspan dev
Xin Long <[email protected]>
sctp: remove useless start_fail from sctp_ht_iter in proc
Haiqing Bai <[email protected]>
tipc: fix the big/little endian issue in tipc_dest
Dexuan Cui <[email protected]>
hv_netvsc: Fix a deadlock by getting rtnl lock earlier in netvsc_probe()
Cong Wang <[email protected]>
tipc: fix a missing rhashtable_walk_exit()
Davide Caratti <[email protected]>
net/sched: act_pedit: fix dump of extended layered op
Michael Chan <[email protected]>
bnxt_en: Do not adjust max_cp_rings by the ones used by RDMA.
Michael Chan <[email protected]>
bnxt_en: Clean up unused functions.
Vlad Buslov <[email protected]>
net: sched: action_ife: take reference to meta module
Cong Wang <[email protected]>
act_ife: fix a potential deadlock
Cong Wang <[email protected]>
act_ife: move tcfa_lock down to where necessary
Alexey Kodanev <[email protected]>
vti6: remove !skb->ignore_df check from vti6_xmit()
Florian Westphal <[email protected]>
tcp: do not restart timewait timer on rst reception
Anthony Wong <[email protected]>
r8169: add support for NCube 8168 network card
Kai-Heng Feng <[email protected]>
r8152: disable RX aggregation on new Dell TB16 dock
Manish Chopra <[email protected]>
qlge: Fix netdev features configuration.
Kees Cook <[email protected]>
net: sched: Fix memory exposure from short TCA_U32_SEL
Anssi Hannula <[email protected]>
net: macb: do not disable MDIO bus at open/close time
Doug Berger <[email protected]>
net: bcmgenet: use MAC link status for fixed phy
Eric Dumazet <[email protected]>
ipv4: tcp: send zero IPID for RST and ACK sent in SYN-RECV and TIME-WAIT state
Cong Wang <[email protected]>
act_ife: fix a potential use-after-free
-------------
Diffstat:
Makefile | 4 +- arch/arm/mach-rockchip/Kconfig | 1 + arch/arm64/Kconfig.platforms | 1 + arch/powerpc/include/asm/topology.h | 5 + arch/powerpc/include/asm/uaccess.h | 13 +- arch/powerpc/kernel/exceptions-64s.S | 6 + arch/powerpc/kernel/smp.c | 5 + arch/powerpc/mm/numa.c | 20 +-- arch/powerpc/platforms/85xx/t1042rdb_diu.c | 4 + arch/powerpc/platforms/pseries/ras.c | 2 +- arch/powerpc/sysdev/mpic_msgr.c | 2 +- arch/riscv/kernel/vdso/Makefile | 4 +- arch/s390/kernel/crash_dump.c | 17 ++- arch/um/Makefile | 3 +- arch/x86/include/asm/mce.h | 1 + arch/x86/include/asm/pgtable-3level.h | 7 +- arch/x86/kernel/tsc.c | 4 +- arch/x86/kvm/mmu.c | 43 ++++++- arch/x86/kvm/vmx.c | 26 ++-- arch/x86/kvm/x86.c | 12 +- arch/x86/xen/mmu_pv.c | 7 +- block/bio.c | 2 +- block/blk-core.c | 4 +- block/blk-mq-tag.c | 3 + block/blk-mq.c | 8 +- block/cfq-iosched.c | 22 ++-- drivers/acpi/acpica/hwregs.c | 9 +- drivers/acpi/scan.c | 5 +- drivers/clk/rockchip/clk-rk3399.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu.h | 6 +- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 23 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 38 ++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 5 + drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 21 +--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h | 4 + drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 17 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 20 ++- drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 2 +- drivers/gpu/drm/amd/amdgpu/psp_v10_0.c | 3 + drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c | 3 +- drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 40 ++++-- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 10 +- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c | 10 +- .../gpu/drm/amd/display/dc/bios/command_table.c | 18 +++ drivers/gpu/drm/amd/display/dc/core/dc_link.c | 11 ++ drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 68 ++++++++--- drivers/gpu/drm/amd/display/dc/dc.h | 1 + .../gpu/drm/amd/display/dc/dce/dce_link_encoder.c | 4 +- .../drm/amd/display/dc/dce100/dce100_resource.c | 2 +- .../drm/amd/display/dc/dce110/dce110_compressor.c | 2 +- .../amd/display/dc/dce110/dce110_hw_sequencer.c | 4 +- .../gpu/drm/amd/display/dc/dce80/dce80_resource.c | 3 + drivers/gpu/drm/amd/display/dc/inc/resource.h | 5 + .../gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c | 43 +++++++ drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c | 5 +- drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c | 2 +- drivers/gpu/drm/drm_edid.c | 6 +- drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 1 + drivers/gpu/drm/i915/i915_drv.c | 10 -- drivers/gpu/drm/i915/i915_drv.h | 8 -- drivers/gpu/drm/i915/i915_reg.h | 1 + drivers/gpu/drm/i915/intel_ddi.c | 4 + drivers/gpu/drm/i915/intel_dp.c | 33 ++--- drivers/gpu/drm/i915/intel_hdmi.c | 8 +- drivers/gpu/drm/i915/intel_lpe_audio.c | 4 +- drivers/gpu/drm/i915/intel_lspcon.c | 2 +- drivers/gpu/drm/i915/intel_lvds.c | 136 +-------------------- drivers/gpu/drm/rockchip/rockchip_drm_vop.c | 69 +++++++---- drivers/gpu/drm/rockchip/rockchip_lvds.c | 4 +- drivers/hid/hid-redragon.c | 26 +--- drivers/i2c/i2c-core-acpi.c | 8 +- drivers/infiniband/hw/hfi1/affinity.c | 24 +++- drivers/infiniband/hw/hns/hns_roce_pd.c | 2 +- drivers/infiniband/hw/hns/hns_roce_qp.c | 5 +- drivers/input/input.c | 16 ++- drivers/iommu/omap-iommu.c | 4 +- drivers/iommu/rockchip-iommu.c | 45 ++++--- drivers/irqchip/irq-bcm7038-l1.c | 4 + drivers/irqchip/irq-stm32-exti.c | 25 ++-- drivers/md/dm-kcopyd.c | 2 + drivers/mfd/sm501.c | 1 + drivers/mtd/ubi/vtbl.c | 20 +-- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 +- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 3 +- drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 7 +- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 20 --- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h | 1 - drivers/net/ethernet/broadcom/genet/bcmgenet.h | 3 + drivers/net/ethernet/broadcom/genet/bcmmii.c | 10 +- drivers/net/ethernet/cadence/macb_main.c | 36 ++++-- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 2 +- .../ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 2 + drivers/net/ethernet/mellanox/mlx5/core/wq.c | 5 +- drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 2 + .../net/ethernet/mellanox/mlxsw/spectrum_router.c | 11 ++ .../ethernet/mellanox/mlxsw/spectrum_switchdev.c | 20 +++ .../net/ethernet/netronome/nfp/nfp_net_common.c | 48 +++++--- drivers/net/ethernet/qlogic/qlge/qlge_main.c | 23 ++-- drivers/net/ethernet/realtek/r8169.c | 7 +- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 +- drivers/net/hyperv/netvsc_drv.c | 16 ++- drivers/net/usb/r8152.c | 4 +- .../broadcom/brcm80211/brcmfmac/cfg80211.c | 8 +- drivers/pci/controller/pci-mvebu.c | 2 +- drivers/pci/probe.c | 12 +- drivers/pinctrl/pinctrl-axp209.c | 26 +++- drivers/platform/x86/asus-nb-wmi.c | 1 + drivers/platform/x86/intel_punit_ipc.c | 1 + drivers/pwm/pwm-meson.c | 3 +- drivers/s390/block/dasd_eckd.c | 10 +- drivers/scsi/aic94xx/aic94xx_init.c | 4 +- drivers/staging/comedi/drivers/ni_mio_common.c | 3 +- drivers/vhost/vhost.c | 2 +- drivers/virtio/virtio_pci_legacy.c | 14 ++- drivers/xen/xen-balloon.c | 2 +- fs/btrfs/check-integrity.c | 7 +- fs/btrfs/dev-replace.c | 6 + fs/btrfs/extent-tree.c | 2 +- fs/btrfs/relocation.c | 23 ++-- fs/btrfs/super.c | 44 ++++--- fs/btrfs/tree-checker.c | 15 ++- fs/btrfs/volumes.c | 94 ++++++++------ fs/cifs/cifs_debug.c | 8 ++ fs/cifs/connect.c | 8 +- fs/cifs/smb2misc.c | 7 ++ fs/cifs/smb2pdu.c | 103 ++++++++-------- fs/dcache.c | 3 +- fs/f2fs/data.c | 8 ++ fs/f2fs/file.c | 48 +++++--- fs/fat/cache.c | 19 +-- fs/fat/fat.h | 5 + fs/fat/fatent.c | 6 +- fs/hfs/brec.c | 7 +- fs/hfsplus/dir.c | 4 +- fs/hfsplus/super.c | 4 +- fs/nfs/nfs4proc.c | 2 +- fs/proc/kcore.c | 4 +- fs/proc/vmcore.c | 2 + fs/reiserfs/reiserfs.h | 2 +- include/linux/pci_ids.h | 2 + include/net/tcp.h | 4 + include/uapi/linux/keyctl.h | 2 +- kernel/bpf/inode.c | 8 +- kernel/bpf/sockmap.c | 120 ++++++++++-------- kernel/fork.c | 5 +- kernel/workqueue.c | 45 ++++--- lib/debugobjects.c | 7 +- mm/Kconfig | 2 +- mm/fadvise.c | 8 +- net/9p/trans_fd.c | 10 +- net/9p/trans_virtio.c | 3 +- net/core/xdp.c | 14 +-- net/ipv4/ip_gre.c | 3 + net/ipv4/tcp_ipv4.c | 6 + net/ipv4/tcp_minisocks.c | 3 +- net/ipv4/tcp_ulp.c | 4 +- net/ipv6/ip6_fib.c | 7 +- net/ipv6/ip6_gre.c | 1 + net/ipv6/ip6_vti.c | 7 +- net/ipv6/netfilter/ip6t_rpfilter.c | 12 +- net/ipv6/route.c | 3 +- net/netfilter/ipvs/ip_vs_core.c | 15 ++- net/netfilter/nf_conntrack_netlink.c | 26 ++-- net/netfilter/nfnetlink_acct.c | 29 ++--- net/netfilter/x_tables.c | 7 +- net/rds/ib_frmr.c | 1 + net/sched/act_ife.c | 78 +++++++----- net/sched/act_pedit.c | 18 ++- net/sched/cls_u32.c | 10 +- net/sctp/proc.c | 8 -- net/sctp/socket.c | 22 ++-- net/sunrpc/auth_gss/gss_krb5_crypto.c | 12 +- net/tipc/name_table.c | 10 +- net/tipc/name_table.h | 9 +- net/tipc/socket.c | 2 + net/tls/tls_main.c | 1 + samples/bpf/xdp_redirect_cpu_user.c | 3 +- samples/bpf/xdp_rxq_info_user.c | 3 +- scripts/coccicheck | 5 +- scripts/depmod.sh | 4 +- scripts/mod/modpost.c | 8 +- security/apparmor/policy_ns.c | 2 +- security/keys/dh.c | 2 +- security/selinux/selinuxfs.c | 33 +++-- sound/soc/codecs/rt5677.c | 2 +- sound/soc/codecs/wm8994.c | 1 + tools/perf/arch/arm64/util/arm-spe.c | 1 + tools/perf/arch/powerpc/util/sym-handling.c | 4 +- tools/perf/util/namespaces.c | 3 + tools/testing/selftests/powerpc/harness.c | 18 ++- 194 files changed, 1498 insertions(+), 954 deletions(-)
Greg Kroah-Hartman
maintainerPATCH 1/197 act_ife: fix a potential use-after-free
4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Cong Wang <[email protected]>
[ Upstream commit 6d784f1625ea68783cc1fb17de8f6cd3e1660c3f ]
Immediately after module_put(), user could delete this module, so e->ops could be already freed before we call e->ops->release().
Fix this by moving module_put() after ops->release().
--- net/sched/act_ife.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/sched/act_ife.c +1 -1
@@ -393,7 +393,6 @@ static void _tcf_ife_cleanup(struct tc_astruct tcf_meta_info *e, *n;list_for_each_entry_safe (e, n, &ife->metalist, metalist) { -module_put (e->ops->owner);list_del (&e->metalist);if (e->metaval) {if (e->ops->release)
@@ -401,6 +400,7 @@ static void _tcf_ife_cleanup(struct tc_aelse kfree (e->metaval); } +module_put (e->ops->owner);kfree (e); } }
Greg Kroah-Hartman
maintainerPATCH 2/197 ipv4: tcp: send zero IPID for RST and ACK sent in SYN-RECV and TIME-WAIT state
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit 431280eebed9f5079553daf003011097763e71fd ]
tcp uses per-cpu (and per namespace) sockets (net->ipv4.tcp_sk) internally to send some control packets.
1) RST packets, through tcp_v4_send_reset()
2) ACK packets in SYN-RECV and TIME-WAIT state, through tcp_v4_send_ack()
These packets assert IP_DF, and also use the hashed IP ident generator to provide an IPv4 ID number.
Geoff Alexander reported this could be used to build off-path attacks.
These packets should not be fragmented, since their size is smaller than IPV4_MIN_MTU. Only some tunneled paths could eventually have to fragment, regardless of inner IPID.
We really can use zero IPID, to address the flaw, and as a bonus, avoid a couple of atomic operations in ip_idents_reserve()
--- net/ipv4/tcp_ipv4.c | 6 ++++++ 1 file changed, 6 insertions(+)
net/ipv4/tcp_ipv4.c +6 -0
@@ -2516,6 +2516,12 @@ static int __net_init tcp_sk_init(structif (res) goto fail;sock_set_flag (sk,SOCK_USE_WRITE_QUEUE ); + +/* Please enforce IP_DF and IPID==0 for RST and +* ACK sent in SYN-RECV and TIME-WAIT state. +*/ +inet_sk (sk)->pmtudisc =IP_PMTUDISC_DO ; + *per_cpu_ptr (net->ipv4.tcp_sk, cpu) = sk; }
Greg Kroah-Hartman
maintainerPATCH 3/197 net: bcmgenet: use MAC link status for fixed phy
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Doug Berger <[email protected]>
[ Upstream commit c3c397c1f16c51601a3fac4fe0c63ad8aa85a904 ]
When using the fixed PHY with GENET (e.g. MOCA) the PHY link status can be determined from the internal link status captured by the MAC. This allows the PHY state machine to use the correct link state with the fixed PHY even if MAC link event interrupts are missed when the net device is opened.
--- drivers/net/ethernet/broadcom/genet/bcmgenet.h | 3 +++ drivers/net/ethernet/broadcom/genet/bcmmii.c | 10 ++++++++-- 2 files changed, 11 insertions(+), 2 deletions(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.h +3 -0
@@ -186,6 +186,9 @@ struct bcmgenet_mib_counters {#define UMAC_MAC1 0x010#define UMAC_MAX_FRAME_LEN 0x014 +#define UMAC_MODE 0x44 +#define MODE_LINK_STATUS (1 << 5) +#define UMAC_EEE_CTRL 0x064#define EN_LPI_RX_PAUSE (1 << 0)#define EN_LPI_TX_PFC (1 << 1)
drivers/net/ethernet/broadcom/genet/bcmmii.c +8 -2
@@ -115,8 +115,14 @@ void bcmgenet_mii_setup(struct net_devicstatic int bcmgenet_fixed_phy_link_update (struct net_device *dev,struct fixed_phy_status *status) { -if (dev && dev->phydev && status) - status->link = dev->phydev->link; +struct bcmgenet_priv *priv; +u32 reg; + +if (dev && dev->phydev && status) { + priv =netdev_priv (dev); + reg =bcmgenet_umac_readl (priv,UMAC_MODE ); + status->link = !!(reg &MODE_LINK_STATUS ); + }return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 4/197 net: macb: do not disable MDIO bus at open/close time
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anssi Hannula <[email protected]>
[ Upstream commit 0da70f808029476001109b6cb076737bc04cea2e ]
macb_reset_hw() is called from macb_close() and indirectly from macb_open(). macb_reset_hw() zeroes the NCR register, including the MPE (Management Port Enable) bit.
This will prevent accessing any other PHYs for other Ethernet MACs on the MDIO bus, which remains registered at macb_reset_hw() time, until macb_init_hw() is called from macb_open() which sets the MPE bit again.
I.e. currently the MDIO bus has a short disruption at open time and is disabled at close time until the interface is opened again.
Fix that by only touching the RE and TE bits when enabling and disabling RX/TX.
v2: Make macb_init_hw() NCR write a single statement.
--- drivers/net/ethernet/cadence/macb_main.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/net/ethernet/cadence/macb_main.c +6 -3
@@ -1957,14 +1957,17 @@ static void macb_reset_hw(struct macb *b {struct macb_queue *queue;unsigned int q; +u32 ctrl =macb_readl (bp,NCR );/* Disable RX and TX (XXX: Should we halt the transmission * more gracefully?) */ -macb_writel (bp,NCR ,0 ); + ctrl &= ~(MACB_BIT (RE ) |MACB_BIT (TE ));/* Clear the stats registers (XXX: Update stats first?) */ -macb_writel (bp,NCR ,MACB_BIT (CLRSTAT )); + ctrl |=MACB_BIT (CLRSTAT ); + +macb_writel (bp,NCR , ctrl);/* Clear all status flags */ macb_writel (bp,TSR ,-1 );
@@ -2152,7 +2155,7 @@ static void macb_init_hw(struct macb *bp }/* Enable TX and RX */ -macb_writel (bp,NCR ,MACB_BIT (RE ) |MACB_BIT (TE ) |MACB_BIT (MPE )); +macb_writel (bp,NCR ,macb_readl (bp,NCR ) |MACB_BIT (RE ) |MACB_BIT (TE )); }/* The hash address register is 64 bits long and takes up two
Greg Kroah-Hartman
maintainerPATCH 5/197 net: sched: Fix memory exposure from short TCA_U32_SEL
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kees Cook <[email protected]>
[ Upstream commit 98c8f125fd8a6240ea343c1aa50a1be9047791b8 ]
Via u32_change(), TCA_U32_SEL has an unspecified type in the netlink policy, so max length isn't enforced, only minimum. This means nkeys (from userspace) was being trusted without checking the actual size of nla_len(), which could lead to a memory over-read, and ultimately an exposure via a call to u32_dump(). Reachability is CAP_NET_ADMIN within a namespace.
--- net/sched/cls_u32.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
net/sched/cls_u32.c +8 -2
@@ -912,6 +912,7 @@ static int u32_change(struct net *net, sstruct nlattr *opt = tca[TCA_OPTIONS ];struct nlattr *tb[TCA_U32_MAX +1 ];u32 htid, flags =0 ; +size_t sel_size;int err;#ifdef CONFIG_CLS_U32_PERF size_t size;
@@ -1074,8 +1075,13 @@ static int u32_change(struct net *net, s } s =nla_data (tb[TCA_U32_SEL ]); + sel_size =struct_size (s, keys, s->nkeys); +if (nla_len (tb[TCA_U32_SEL ]) < sel_size) { + err = -EINVAL ; + goto erridr; + } - n =kzalloc (sizeof (*n) + s->nkeys*sizeof (struct tc_u32_key ),GFP_KERNEL ); + n = kzalloc(offsetof(typeof(*n), sel) + sel_size, GFP_KERNEL);if (n ==NULL ) { err = -ENOBUFS ; goto erridr;
@@ -1090,7 +1096,7 @@ static int u32_change(struct net *net, s }#endif -memcpy (&n->sel, s,sizeof (*s) + s->nkeys*sizeof (struct tc_u32_key )); +memcpy (&n->sel, s, sel_size);RCU_INIT_POINTER (n->ht_up, ht); n->handle = handle; n->fshift = s->hmask ?ffs (ntohl (s->hmask)) -1 :0 ;
Greg Kroah-Hartman
maintainerPATCH 6/197 qlge: Fix netdev features configuration.
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Manish Chopra <[email protected]>
[ Upstream commit 6750c87074c5b534d82fdaabb1deb45b8f1f57de ]
qlge_fix_features() is not supposed to modify hardware or driver state, rather it is supposed to only fix requested fetures bits. Currently qlge_fix_features() also goes for interface down and up unnecessarily if there is not even any change in features set.
This patch changes/fixes following -
1) Move reload of interface or device re-config from qlge_fix_features() to qlge_set_features().
2) Reload of interface in qlge_set_features() only if relevant feature bit (NETIF_F_HW_VLAN_CTAG_RX) is changed.
3) Get rid of qlge_fix_features() since driver is not really required to fix any features bit.
--- drivers/net/ethernet/qlogic/qlge/qlge_main.c | 23 ++++++++--------------- 1 file changed, 8 insertions(+), 15 deletions(-)
drivers/net/ethernet/qlogic/qlge/qlge_main.c +8 -15
@@ -2384,26 +2384,20 @@ static int qlge_update_hw_vlan_features(return status; } -static netdev_features_t qlge_fix_features (struct net_device *ndev, -netdev_features_t features) -{ -int err; - -/* Update the behavior of vlan accel in the adapter */ - err =qlge_update_hw_vlan_features (ndev, features); -if (err) -return err; - -return features; -} -static int qlge_set_features (struct net_device *ndev,netdev_features_t features) {netdev_features_t changed = ndev->features ^ features; +int err; + +if (changed &NETIF_F_HW_VLAN_CTAG_RX ) { +/* Update the behavior of vlan accel in the adapter */ + err =qlge_update_hw_vlan_features (ndev, features); +if (err) +return err; -if (changed &NETIF_F_HW_VLAN_CTAG_RX )qlge_vlan_mode (ndev, features); + }return 0 ; }
@@ -4719,7 +4713,6 @@ static const struct net_device_ops qlge_ .ndo_set_mac_address = qlge_set_mac_address, .ndo_validate_addr = eth_validate_addr, .ndo_tx_timeout = qlge_tx_timeout, - .ndo_fix_features = qlge_fix_features, .ndo_set_features = qlge_set_features, .ndo_vlan_rx_add_vid = qlge_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = qlge_vlan_rx_kill_vid,
Greg Kroah-Hartman
maintainerPATCH 7/197 r8152: disable RX aggregation on new Dell TB16 dock
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kai-Heng Feng <[email protected]>
[ Upstream commit 176eb614b118c96e7797f5ddefd10708c316f621 ]
There's a new Dell TB16 dock with a different iSerialNumber.
Apply the same fix from commit 0b1655143df0 ("r8152: disable RX aggregation on Dell TB16 dock") to this model.
--- drivers/net/usb/r8152.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/usb/r8152.c +2 -2
@@ -5214,8 +5214,8 @@ static int rtl8152_probe(struct usb_inte netdev->hw_features &= ~NETIF_F_RXCSUM ; } -if (le16_to_cpu (udev->descriptor.bcdDevice) ==0x3011 && - udev->serial && !strcmp (udev->serial,"000001000000" )) { +if (le16_to_cpu (udev->descriptor.bcdDevice) ==0x3011 && udev->serial && + (!strcmp (udev->serial,"000001000000" ) || !strcmp (udev->serial,"000002000000" ))) {dev_info (&udev->dev,"Dell TB16 Dock, disable RX aggregation" );set_bit (DELL_TB_RX_AGG_BUG , &tp->flags); }
Greg Kroah-Hartman
maintainerPATCH 8/197 r8169: add support for NCube 8168 network card
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anthony Wong <[email protected]>
[ Upstream commit 9fd0e09a4e86499639653243edfcb417a05c5c46 ]
This card identifies itself as:
Ethernet controller [0200]: NCube Device [10ff:8168] (rev 06)
Subsystem: TP-LINK Technologies Co., Ltd. Device [7470:3468]
Adding a new entry to rtl8169_pci_tbl makes the card work.
--- drivers/net/ethernet/realtek/r8169.c | 1 + include/linux/pci_ids.h | 2 ++ 2 files changed, 3 insertions(+)
drivers/net/ethernet/realtek/r8169.c +1 -0
@@ -303,6 +303,7 @@ static const struct pci_device_id rtl816 { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8161), 0, 0, RTL_CFG_1 }, { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8167), 0, 0, RTL_CFG_0 }, { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8168), 0, 0, RTL_CFG_1 }, + { PCI_DEVICE(PCI_VENDOR_ID_NCUBE, 0x8168), 0, 0, RTL_CFG_1 }, { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8169), 0, 0, RTL_CFG_0 }, { PCI_VENDOR_ID_DLINK, 0x4300, PCI_VENDOR_ID_DLINK, 0x4b10, 0, 0, RTL_CFG_1 },
include/linux/pci_ids.h +2 -0
@@ -3082,4 +3082,6 @@#define PCI_VENDOR_ID_OCZ 0x1b85 +#define PCI_VENDOR_ID_NCUBE 0x10ff +#endif /* _LINUX_PCI_IDS_H */
Greg Kroah-Hartman
maintainerPATCH 9/197 tcp: do not restart timewait timer on rst reception
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Westphal <[email protected]>
[ Upstream commit 63cc357f7bba6729869565a12df08441a5995d9a ]
RFC 1337 says:
''Ignore RST segments in TIME-WAIT state.
If the 2 minute MSL is enforced, this fix avoids all three hazards.''
So with net.ipv4.tcp_rfc1337=1, expected behaviour is to have TIME-WAIT sk expire rather than removing it instantly when a reset is received.
However, Linux will also re-start the TIME-WAIT timer.
This causes connect to fail when tying to re-use ports or very long delays (until syn retry interval exceeds MSL).
packetdrill test case:
// Demonstrate bogus rearming of TIME-WAIT timer in rfc1337 mode. sysctl net.ipv4.tcp_rfc1337=1
0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 0.000 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 0.000 bind(3, ..., ...) = 0
0.000 listen(3, 1) = 0
0.100 < S 0:0(0) win 29200 <mss 1460,nop,nop,sackOK,nop,wscale 7> 0.100 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 7> 0.200 < . 1:1(0) ack 1 win 257
0.200 accept(3, ..., ...) = 4
// Receive first segment
0.310 < P. 1:1001(1000) ack 1 win 46
// Send one ACK
0.310 > . 1:1(0) ack 1001
// read 1000 byte
0.310 read(4, ..., 1000) = 1000
// Application writes 100 bytes
0.350 write(4, ..., 100) = 100
0.350 > P. 1:101(100) ack 1001
// ACK
0.500 < . 1001:1001(0) ack 101 win 257
// close the connection
0.600 close(4) = 0
0.600 > F. 101:101(0) ack 1001 win 244
// Our side is in FIN_WAIT_1 & waits for ack to fin 0.7 < . 1001:1001(0) ack 102 win 244
// Our side is in FIN_WAIT_2 with no outstanding data. 0.8 < F. 1001:1001(0) ack 102 win 244
0.8 > . 102:102(0) ack 1002 win 244
// Our side is now in TIME_WAIT state, send ack for fin. 0.9 < F. 1002:1002(0) ack 102 win 244
0.9 > . 102:102(0) ack 1002 win 244
// Peer reopens with in-window SYN:
1.000 < S 1000:1000(0) win 9200 <mss 1460,nop,nop,sackOK,nop,wscale 7>
// Therefore, reply with ACK.
1.000 > . 102:102(0) ack 1002 win 244
// Peer sends RST for this ACK. Normally this RST results // in tw socket removal, but rfc1337=1 setting prevents this. 1.100 < R 1002:1002(0) win 244
// second syn. Due to rfc1337=1 expect another pure ACK. 31.0 < S 1000:1000(0) win 9200 <mss 1460,nop,nop,sackOK,nop,wscale 7> 31.0 > . 102:102(0) ack 1002 win 244
// .. and another RST from peer.
31.1 < R 1002:1002(0) win 244
31.2 echo no timer restart;ss -m -e -a -i -n -t -o state TIME-WAIT
// third syn after one minute. Time-Wait socket should have expired by now. 63.0 < S 1000:1000(0) win 9200 <mss 1460,nop,nop,sackOK,nop,wscale 7>
// so we expect a syn-ack & 3whs to proceed from here on. 63.0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 7>
Without this patch, 'ss' shows restarts of tw timer and last packet is thus just another pure ack, more than one minute later.
This restores the original code from commit 283fd6cf0be690a83 ("Merge in ANK networking jumbo patch") in netdev-vger-cvs.git .
For some reason the else branch was removed/lost in 1f28b683339f7 ("Merge in TCP/UDP optimizations and [..]") and timer restart became unconditional.
--- net/ipv4/tcp_minisocks.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/ipv4/tcp_minisocks.c +2 -1
@@ -184,8 +184,9 @@ kill:inet_twsk_deschedule_put (tw);return TCP_TW_SUCCESS ; } + }else { +inet_twsk_reschedule (tw,TCP_TIMEWAIT_LEN ); } -inet_twsk_reschedule (tw,TCP_TIMEWAIT_LEN );if (tmp_opt.saw_tstamp) { tcptw->tw_ts_recent = tmp_opt.rcv_tsval;
Greg Kroah-Hartman
maintainerPATCH 10/197 vti6: remove !skb->ignore_df check from vti6_xmit()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexey Kodanev <[email protected]>
[ Upstream commit 9f2895461439fda2801a7906fb4c5fb3dbb37a0a ]
Before the commit d6990976af7c ("vti6: fix PMTU caching and reporting on xmit") '!skb->ignore_df' check was always true because the function skb_scrub_packet() was called before it, resetting ignore_df to zero.
In the commit, skb_scrub_packet() was moved below, and now this check can be false for the packet, e.g. when sending it in the two fragments, this prevents successful PMTU updates in such case. The next attempts to send the packet lead to the same tx error. Moreover, vti6 initial MTU value relies on PMTU adjustments.
This issue can be reproduced with the following LTP test script:
udp_ipsec_vti.sh -6 -p ah -m tunnel -s 2000
--- net/ipv6/ip6_vti.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/ip6_vti.c +1 -1
@@ -481,7 +481,7 @@ vti6_xmit(struct sk_buff *skb, struct ne } mtu =dst_mtu (dst); -if (!skb->ignore_df && skb->len > mtu) { +if (skb->len > mtu) {skb_dst_update_pmtu (skb, mtu);if (skb->protocol ==htons (ETH_P_IPV6 )) {
Greg Kroah-Hartman
maintainerPATCH 11/197 act_ife: move tcfa_lock down to where necessary
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Cong Wang <[email protected]>
[ Upstream commit 4e407ff5cd67ec76eeeea1deec227b7982dc7f66 ]
The only time we need to take tcfa_lock is when adding a new metainfo to an existing ife->metalist. We don't need to take tcfa_lock so early and so broadly in tcf_ife_init().
This means we can always take ife_mod_lock first, avoid the reverse locking ordering warning as reported by Vlad.
--- net/sched/act_ife.c | 36 +++++++++++++----------------------- 1 file changed, 13 insertions(+), 23 deletions(-)
net/sched/act_ife.c +13 -23
@@ -265,10 +265,8 @@ static const char *ife_meta_id2name(u32#endif /* called when adding new meta information -* under ife->tcf_lock for existing action */ -static int load_metaops_and_vet (struct tcf_ife_info *ife,u32 metaid, -void *val,int len,bool exists) +static int load_metaops_and_vet (u32 metaid,void *val,int len) {struct tcf_meta_ops *ops =find_ife_oplist (metaid);int ret =0 ;
@@ -276,13 +274,9 @@ static int load_metaops_and_vet(struct tif (!ops) { ret = -ENOENT ;#ifdef CONFIG_MODULES -if (exists) -spin_unlock_bh (&ife->tcf_lock);rtnl_unlock ();request_module ("ife-meta-%s" ,ife_meta_id2name (metaid));rtnl_lock (); -if (exists) -spin_lock_bh (&ife->tcf_lock); ops =find_ife_oplist (metaid);#endif }
@@ -299,10 +293,9 @@ static int load_metaops_and_vet(struct t }/* called when adding new meta information -* under ife->tcf_lock for existing action */ static int add_metainfo (struct tcf_ife_info *ife,u32 metaid,void *metaval, -int len,bool atomic) +int len,bool atomic,bool exists) {struct tcf_meta_info *mi =NULL ;struct tcf_meta_ops *ops =find_ife_oplist (metaid);
@@ -329,12 +322,16 @@ static int add_metainfo(struct tcf_ife_i } } +if (exists) +spin_lock_bh (&ife->tcf_lock);list_add_tail (&mi->metalist, &ife->metalist); +if (exists) +spin_unlock_bh (&ife->tcf_lock);return ret; } -static int use_all_metadata (struct tcf_ife_info *ife) +static int use_all_metadata (struct tcf_ife_info *ife,bool exists) {struct tcf_meta_ops *o;int rc =0 ;
@@ -342,7 +339,7 @@ static int use_all_metadata(struct tcf_iread_lock (&ife_mod_lock);list_for_each_entry (o, &ifeoplist, list) { - rc =add_metainfo (ife, o->metaid,NULL ,0 , true); + rc =add_metainfo (ife, o->metaid,NULL ,0 , true, exists);if (rc ==0 ) installed +=1 ; }
@@ -419,7 +416,6 @@ static void tcf_ife_cleanup(struct tc_ackfree_rcu (p, rcu); } -/* under ife->tcf_lock for existing action */ static int populate_metalist (struct tcf_ife_info *ife,struct nlattr **tb,bool exists) {
@@ -433,11 +429,11 @@ static int populate_metalist(struct tcf_ val =nla_data (tb[i]); len =nla_len (tb[i]); - rc =load_metaops_and_vet (ife, i, val, len, exists); + rc =load_metaops_and_vet (i, val, len);if (rc !=0 )return rc; - rc =add_metainfo (ife, i, val, len, exists); + rc =add_metainfo (ife, i, val, len, false, exists);if (rc)return rc; }
@@ -531,8 +527,6 @@ static int tcf_ife_init(struct net *net, p->eth_type = ife_type; } -if (exists) -spin_lock_bh (&ife->tcf_lock);if (ret ==ACT_P_CREATED )INIT_LIST_HEAD (&ife->metalist);
@@ -544,9 +538,6 @@ static int tcf_ife_init(struct net *net, metadata_parse_err:if (ret ==ACT_P_CREATED )tcf_idr_release (*a, bind); - -if (exists) -spin_unlock_bh (&ife->tcf_lock);kfree (p);return err; }
@@ -561,18 +552,17 @@ metadata_parse_err:* as we can. You better have at least one else we are * going to bail out */ - err =use_all_metadata (ife); + err =use_all_metadata (ife, exists);if (err) {if (ret ==ACT_P_CREATED )tcf_idr_release (*a, bind); - -if (exists) -spin_unlock_bh (&ife->tcf_lock);kfree (p);return err; } } +if (exists) +spin_lock_bh (&ife->tcf_lock); ife->tcf_action = parm->action;if (exists)spin_unlock_bh (&ife->tcf_lock);
Greg Kroah-Hartman
maintainerPATCH 12/197 act_ife: fix a potential deadlock
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Cong Wang <[email protected]>
[ Upstream commit 5ffe57da29b3802baeddaa40909682bbb4cb4d48 ]
use_all_metadata() acquires read_lock(&ife_mod_lock), then calls add_metainfo() which calls find_ife_oplist() which acquires the same lock again. Deadlock!
Introduce __add_metainfo() which accepts struct tcf_meta_ops *ops as an additional parameter and let its callers to decide how to find it. For use_all_metadata(), it already has ops, no need to find it again, just call __add_metainfo() directly.
And, as ife_mod_lock is only needed for find_ife_oplist(), this means we can make non-atomic allocation for populate_metalist() now.
--- net/sched/act_ife.c | 34 +++++++++++++++++++++------------- 1 file changed, 21 insertions(+), 13 deletions(-)
net/sched/act_ife.c +21 -13
@@ -294,22 +294,16 @@ static int load_metaops_and_vet(u32 meta/* called when adding new meta information */ -static int add_metainfo (struct tcf_ife_info *ife,u32 metaid,void *metaval, -int len,bool atomic,bool exists) +static int __add_metainfo (const struct tcf_meta_ops *ops, +struct tcf_ife_info *ife,u32 metaid,void *metaval, +int len,bool atomic,bool exists) {struct tcf_meta_info *mi =NULL ; -struct tcf_meta_ops *ops =find_ife_oplist (metaid);int ret =0 ; -if (!ops) -return -ENOENT ; - mi =kzalloc (sizeof (*mi), atomic ?GFP_ATOMIC :GFP_KERNEL ); -if (!mi) { -/*put back what find_ife_oplist took */ -module_put (ops->owner); +if (!mi)return -ENOMEM ; - } mi->metaid = metaid; mi->ops = ops;
@@ -317,7 +311,6 @@ static int add_metainfo(struct tcf_ife_i ret = ops->alloc (mi, metaval, atomic ?GFP_ATOMIC :GFP_KERNEL );if (ret !=0 ) {kfree (mi); -module_put (ops->owner);return ret; } }
@@ -331,6 +324,21 @@ static int add_metainfo(struct tcf_ife_ireturn ret; } +static int add_metainfo (struct tcf_ife_info *ife,u32 metaid,void *metaval, +int len,bool exists) +{ +const struct tcf_meta_ops *ops =find_ife_oplist (metaid); +int ret; + +if (!ops) +return -ENOENT ; + ret =__add_metainfo (ops, ife, metaid, metaval, len, false, exists); +if (ret) +/*put back what find_ife_oplist took */ +module_put (ops->owner); +return ret; +} +static int use_all_metadata (struct tcf_ife_info *ife,bool exists) {struct tcf_meta_ops *o;
@@ -339,7 +347,7 @@ static int use_all_metadata(struct tcf_iread_lock (&ife_mod_lock);list_for_each_entry (o, &ifeoplist, list) { - rc =add_metainfo (ife, o->metaid,NULL ,0 , true, exists); + rc =__add_metainfo (o, ife, o->metaid,NULL ,0 , true, exists);if (rc ==0 ) installed +=1 ; }
@@ -433,7 +441,7 @@ static int populate_metalist(struct tcf_if (rc !=0 )return rc; - rc =add_metainfo (ife, i, val, len, false, exists); + rc =add_metainfo (ife, i, val, len, exists);if (rc)return rc; }
Greg Kroah-Hartman
maintainerPATCH 13/197 net: sched: action_ife: take reference to meta module
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vlad Buslov <[email protected]>
[ Upstream commit 84cb8eb26cb9ce3c79928094962a475a9d850a53 ]
Recent refactoring of add_metainfo() caused use_all_metadata() to add metainfo to ife action metalist without taking reference to module. This causes warning in module_put called from ife action cleanup function.
Implement add_metainfo_and_get_ops() function that returns with reference to module taken if metainfo was added successfully, and call it from use_all_metadata(), instead of calling __add_metainfo() directly.
Example warning:
[ 646.344393] WARNING: CPU: 1 PID: 2278 at kernel/module.c:1139 module_put+0x1cb/0x230 [ 646.352437] Modules linked in: act_meta_skbtcindex act_meta_mark act_meta_skbprio act_ife ife veth nfsv3 nfs fscache xt_CHECKSUM iptable_mangle ipt_MASQUERADE iptable_nat nf_nat_ipv4 nf_nat xt_conntrack nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 libcrc32c tun ebtable_filter ebtables ip6table_filter ip6_tables bridge stp llc mlx5_ib ib_uverbs ib_core intel_rapl sb_edac x86_pkg_temp_thermal mlx5_core coretemp kvm_intel kvm nfsd igb irqbypass crct10dif_pclmul devlink crc32_pclmul mei_me joydev ses crc32c_intel enclosure auth_rpcgss i2c_algo_bit ioatdma ptp mei pps_core ghash_clmulni_intel iTCO_wdt iTCO_vendor_support pcspkr dca ipmi_ssif lpc_ich target_core_mod i2c_i801 ipmi_si ipmi_devintf pcc_cpufreq wmi ipmi_msghandler nfs_acl lockd acpi_pad acpi_power_meter grace sunrpc mpt3sas raid_class scsi_transport_sas [ 646.425631] CPU: 1 PID: 2278 Comm: tc Not tainted 4.19.0-rc1+ #799 [ 646.432187] Hardware name: Supermicro SYS-2028TP-DECR/X10DRT-P, BIOS 2.0b 03/30/2017 [ 646.440595] RIP: 0010:module_put+0x1cb/0x230 [ 646.445238] Code: f3 66 94 02 e8 26 ff fa ff 85 c0 74 11 0f b6 1d 51 30 94 02 80 fb 01 77 60 83 e3 01 74 13 65 ff 0d 3a 83 db 73 e9 2b ff ff ff <0f> 0b e9 00 ff ff ff e8 59 01 fb ff 85 c0 75 e4 48 c7 c2 20 62 6b [ 646.464997] RSP: 0018:ffff880354d37068 EFLAGS: 00010286 [ 646.470599] RAX: 0000000000000000 RBX: ffffffffc0a52518 RCX: ffffffff8c2668db [ 646.478118] RDX: 0000000000000003 RSI: dffffc0000000000 RDI: ffffffffc0a52518 [ 646.485641] RBP: ffffffffc0a52180 R08: fffffbfff814a4a4 R09: fffffbfff814a4a3 [ 646.493164] R10: ffffffffc0a5251b R11: fffffbfff814a4a4 R12: 1ffff1006a9a6e0d [ 646.500687] R13: 00000000ffffffff R14: ffff880362bab890 R15: dead000000000100 [ 646.508213] FS: 00007f4164c99800(0000) GS:ffff88036fe40000(0000) knlGS:0000000000000000 [ 646.516961] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 646.523080] CR2: 00007f41638b8420 CR3: 0000000351df0004 CR4: 00000000001606e0 [ 646.530595] Call Trace: [ 646.533408] ? find_symbol_in_section+0x260/0x260 [ 646.538509] tcf_ife_cleanup+0x11b/0x200 [act_ife] [ 646.543695] tcf_action_cleanup+0x29/0xa0 [ 646.548078] __tcf_action_put+0x5a/0xb0 [ 646.552289] ? nla_put+0x65/0xe0 [ 646.555889] __tcf_idr_release+0x48/0x60 [ 646.560187] tcf_generic_walker+0x448/0x6b0 [ 646.564764] ? tcf_action_dump_1+0x450/0x450 [ 646.569411] ? __lock_is_held+0x84/0x110 [ 646.573720] ? tcf_ife_walker+0x10c/0x20f [act_ife] [ 646.578982] tca_action_gd+0x972/0xc40 [ 646.583129] ? tca_get_fill.constprop.17+0x250/0x250 [ 646.588471] ? mark_lock+0xcf/0x980 [ 646.592324] ? check_chain_key+0x140/0x1f0 [ 646.596832] ? debug_show_all_locks+0x240/0x240 [ 646.601839] ? memset+0x1f/0x40 [ 646.605350] ? nla_parse+0xca/0x1a0 [ 646.609217] tc_ctl_action+0x215/0x230 [ 646.613339] ? tcf_action_add+0x220/0x220 [ 646.617748] rtnetlink_rcv_msg+0x56a/0x6d0 [ 646.622227] ? rtnl_fdb_del+0x3f0/0x3f0 [ 646.626466] netlink_rcv_skb+0x18d/0x200 [ 646.630752] ? rtnl_fdb_del+0x3f0/0x3f0 [ 646.634959] ? netlink_ack+0x500/0x500 [ 646.639106] netlink_unicast+0x2d0/0x370 [ 646.643409] ? netlink_attachskb+0x340/0x340 [ 646.648050] ? _copy_from_iter_full+0xe9/0x3e0 [ 646.652870] ? import_iovec+0x11e/0x1c0 [ 646.657083] netlink_sendmsg+0x3b9/0x6a0 [ 646.661388] ? netlink_unicast+0x370/0x370 [ 646.665877] ? netlink_unicast+0x370/0x370 [ 646.670351] sock_sendmsg+0x6b/0x80 [ 646.674212] ___sys_sendmsg+0x4a1/0x520 [ 646.678443] ? copy_msghdr_from_user+0x210/0x210 [ 646.683463] ? lock_downgrade+0x320/0x320 [ 646.687849] ? debug_show_all_locks+0x240/0x240 [ 646.692760] ? do_raw_spin_unlock+0xa2/0x130 [ 646.697418] ? _raw_spin_unlock+0x24/0x30 [ 646.701798] ? __handle_mm_fault+0x1819/0x1c10 [ 646.706619] ? __pmd_alloc+0x320/0x320 [ 646.710738] ? debug_show_all_locks+0x240/0x240 [ 646.715649] ? restore_nameidata+0x7b/0xa0 [ 646.720117] ? check_chain_key+0x140/0x1f0 [ 646.724590] ? check_chain_key+0x140/0x1f0 [ 646.729070] ? __fget_light+0xbc/0xd0 [ 646.733121] ? __sys_sendmsg+0xd7/0x150 [ 646.737329] __sys_sendmsg+0xd7/0x150 [ 646.741359] ? __ia32_sys_shutdown+0x30/0x30 [ 646.746003] ? up_read+0x53/0x90 [ 646.749601] ? __do_page_fault+0x484/0x780 [ 646.754105] ? do_syscall_64+0x1e/0x2c0 [ 646.758320] do_syscall_64+0x72/0x2c0 [ 646.762353] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 646.767776] RIP: 0033:0x7f4163872150 [ 646.771713] Code: 8b 15 3c 7d 2b 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb cd 66 0f 1f 44 00 00 83 3d b9 d5 2b 00 00 75 10 b8 2e 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 be cd 00 00 48 89 04 24 [ 646.791474] RSP: 002b:00007ffdef7d6b58 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 646.799721] RAX: ffffffffffffffda RBX: 0000000000000024 RCX: 00007f4163872150 [ 646.807240] RDX: 0000000000000000 RSI: 00007ffdef7d6bd0 RDI: 0000000000000003 [ 646.814760] RBP: 000000005b8b9482 R08: 0000000000000001 R09: 0000000000000000 [ 646.822286] R10: 00000000000005e7 R11: 0000000000000246 R12: 00007ffdef7dad20 [ 646.829807] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000679bc0 [ 646.837360] irq event stamp: 6083 [ 646.841043] hardirqs last enabled at (6081): [<ffffffff8c220a7d>] __call_rcu+0x17d/0x500 [ 646.849882] hardirqs last disabled at (6083): [<ffffffff8c004f06>] trace_hardirqs_off_thunk+0x1a/0x1c [ 646.859775] softirqs last enabled at (5968): [<ffffffff8d4004a1>] __do_softirq+0x4a1/0x6ee [ 646.868784] softirqs last disabled at (6082): [<ffffffffc0a78759>] tcf_ife_cleanup+0x39/0x200 [act_ife] [ 646.878845] ---[ end trace b1b8c12ffe51e657 ]---
--- net/sched/act_ife.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-)
net/sched/act_ife.c +15 -1
@@ -324,6 +324,20 @@ static int __add_metainfo(const struct treturn ret; } +static int add_metainfo_and_get_ops (const struct tcf_meta_ops *ops, +struct tcf_ife_info *ife,u32 metaid, +bool exists) +{ +int ret; + +if (!try_module_get (ops->owner)) +return -ENOENT ; + ret =__add_metainfo (ops, ife, metaid,NULL ,0 , true, exists); +if (ret) +module_put (ops->owner); +return ret; +} +static int add_metainfo (struct tcf_ife_info *ife,u32 metaid,void *metaval,int len,bool exists) {
@@ -347,7 +361,7 @@ static int use_all_metadata(struct tcf_iread_lock (&ife_mod_lock);list_for_each_entry (o, &ifeoplist, list) { - rc =__add_metainfo (o, ife, o->metaid,NULL ,0 , true, exists); + rc =add_metainfo_and_get_ops (o, ife, o->metaid, exists);if (rc ==0 ) installed +=1 ; }
Greg Kroah-Hartman
maintainerPATCH 14/197 bnxt_en: Clean up unused functions.
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michael Chan <[email protected]>
[ Upstream commit ad95c27bdb930105f3eea02621bda157caf2862d ]
Remove unused bnxt_subtract_ulp_resources(). Change bnxt_get_max_func_irqs() to static since it is only locally used.
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 - drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 15 --------------- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h | 1 - 4 files changed, 1 insertion(+), 18 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +1 -1
@@ -5912,7 +5912,7 @@ void bnxt_set_max_func_cp_rings(struct b bp->hw_resc.max_cp_rings = max; } -unsigned int bnxt_get_max_func_irqs (struct bnxt *bp) +static unsigned int bnxt_get_max_func_irqs (struct bnxt *bp) {struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
drivers/net/ethernet/broadcom/bnxt/bnxt.h +0 -1
@@ -1469,7 +1469,6 @@ unsigned int bnxt_get_max_func_stat_ctxsvoid bnxt_set_max_func_stat_ctxs (struct bnxt *bp,unsigned int max);unsigned int bnxt_get_max_func_cp_rings (struct bnxt *bp);void bnxt_set_max_func_cp_rings (struct bnxt *bp,unsigned int max); -unsigned int bnxt_get_max_func_irqs (struct bnxt *bp);int bnxt_get_avail_msix (struct bnxt *bp,int num);int bnxt_reserve_rings (struct bnxt *bp);void bnxt_tx_disable (struct bnxt *bp);
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +0 -15
@@ -220,21 +220,6 @@ int bnxt_get_ulp_msix_base(struct bnxt *return 0 ; } -void bnxt_subtract_ulp_resources (struct bnxt *bp,int ulp_id) -{ -ASSERT_RTNL (); -if (bnxt_ulp_registered (bp->edev, ulp_id)) { -struct bnxt_en_dev *edev = bp->edev; -unsigned int msix_req, max; - - msix_req = edev->ulp_tbl[ulp_id].msix_requested; - max =bnxt_get_max_func_cp_rings (bp); -bnxt_set_max_func_cp_rings (bp, max - msix_req); - max =bnxt_get_max_func_stat_ctxs (bp); -bnxt_set_max_func_stat_ctxs (bp, max -1 ); - } -} -static int bnxt_send_msg (struct bnxt_en_dev *edev,int ulp_id,struct bnxt_fw_msg *fw_msg) {
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h +0 -1
@@ -90,7 +90,6 @@ static inline bool bnxt_ulp_registered(sint bnxt_get_ulp_msix_num (struct bnxt *bp);int bnxt_get_ulp_msix_base (struct bnxt *bp); -void bnxt_subtract_ulp_resources (struct bnxt *bp,int ulp_id);void bnxt_ulp_stop (struct bnxt *bp);void bnxt_ulp_start (struct bnxt *bp);void bnxt_ulp_sriov_cfg (struct bnxt *bp,int num_vfs);
Greg Kroah-Hartman
maintainerPATCH 15/197 bnxt_en: Do not adjust max_cp_rings by the ones used by RDMA.
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michael Chan <[email protected]>
[ Upstream commit 00fe9c326d2027f2437dea38ef0e82f9d02d94c0 ]
Currently, the driver adjusts the bp->hw_resc.max_cp_rings by the number of MSIX vectors used by RDMA. There is one code path in open that needs to check the true max_cp_rings including any used by RDMA. This code is now checking for the reduced max_cp_rings which will fail when the number of cp rings is very small.
To fix this in a clean way, we don't adjust max_cp_rings anymore. Instead, we add a helper bnxt_get_max_func_cp_rings_for_en() to get the reduced max_cp_rings when appropriate.
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 7 ++++--- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 7 ++++--- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 5 ----- 4 files changed, 9 insertions(+), 12 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +4 -3
@@ -5907,9 +5907,9 @@ unsigned int bnxt_get_max_func_cp_rings(return bp->hw_resc.max_cp_rings; } -void bnxt_set_max_func_cp_rings (struct bnxt *bp,unsigned int max) +unsigned int bnxt_get_max_func_cp_rings_for_en (struct bnxt *bp) { - bp->hw_resc.max_cp_rings = max; +return bp->hw_resc.max_cp_rings -bnxt_get_ulp_msix_num (bp); }static unsigned int bnxt_get_max_func_irqs (struct bnxt *bp)
@@ -8492,7 +8492,8 @@ static void _bnxt_get_max_rings(struct b *max_tx = hw_resc->max_tx_rings; *max_rx = hw_resc->max_rx_rings; - *max_cp =min_t (int, hw_resc->max_irqs, hw_resc->max_cp_rings); + *max_cp =min_t (int,bnxt_get_max_func_cp_rings_for_en (bp), + hw_resc->max_irqs); *max_cp =min_t (int, *max_cp, hw_resc->max_stat_ctxs); max_ring_grps = hw_resc->max_hw_ring_grps;if (BNXT_CHIP_TYPE_NITRO_A0 (bp) &&BNXT_PF (bp)) {
drivers/net/ethernet/broadcom/bnxt/bnxt.h +1 -1
@@ -1468,7 +1468,7 @@ int bnxt_hwrm_set_coal(struct bnxt *);unsigned int bnxt_get_max_func_stat_ctxs (struct bnxt *bp);void bnxt_set_max_func_stat_ctxs (struct bnxt *bp,unsigned int max);unsigned int bnxt_get_max_func_cp_rings (struct bnxt *bp); -void bnxt_set_max_func_cp_rings (struct bnxt *bp,unsigned int max); +unsigned int bnxt_get_max_func_cp_rings_for_en (struct bnxt *bp);int bnxt_get_avail_msix (struct bnxt *bp,int num);int bnxt_reserve_rings (struct bnxt *bp);void bnxt_tx_disable (struct bnxt *bp);
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +4 -3
@@ -451,7 +451,7 @@ static int bnxt_hwrm_func_vf_resc_cfg(stbnxt_hwrm_cmd_hdr_init (bp, &req,HWRM_FUNC_VF_RESOURCE_CFG ,-1 ,-1 ); - vf_cp_rings = hw_resc->max_cp_rings - bp->cp_nr_rings; + vf_cp_rings =bnxt_get_max_func_cp_rings_for_en (bp) - bp->cp_nr_rings; vf_stat_ctx = hw_resc->max_stat_ctxs - bp->num_stat_ctxs;if (bp->flags &BNXT_FLAG_AGG_RINGS ) vf_rx_rings = hw_resc->max_rx_rings - bp->rx_nr_rings *2 ;
@@ -544,7 +544,8 @@ static int bnxt_hwrm_func_cfg(struct bnx max_stat_ctxs = hw_resc->max_stat_ctxs;/* Remaining rings are distributed equally amongs VF's for now */ - vf_cp_rings = (hw_resc->max_cp_rings - bp->cp_nr_rings) / num_vfs; + vf_cp_rings = (bnxt_get_max_func_cp_rings_for_en (bp) - + bp->cp_nr_rings) / num_vfs; vf_stat_ctx = (max_stat_ctxs - bp->num_stat_ctxs) / num_vfs;if (bp->flags &BNXT_FLAG_AGG_RINGS ) vf_rx_rings = (hw_resc->max_rx_rings - bp->rx_nr_rings * 2) /
@@ -638,7 +639,7 @@ static int bnxt_sriov_enable(struct bnxt*/ vfs_supported = *num_vfs; - avail_cp = hw_resc->max_cp_rings - bp->cp_nr_rings; + avail_cp =bnxt_get_max_func_cp_rings_for_en (bp) - bp->cp_nr_rings; avail_stat = hw_resc->max_stat_ctxs - bp->num_stat_ctxs; avail_cp =min_t (int, avail_cp, avail_stat);
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +0 -5
@@ -169,7 +169,6 @@ static int bnxt_req_msix_vecs(struct bnx edev->ulp_tbl[ulp_id].msix_requested = avail_msix; }bnxt_fill_msix_vecs (bp, ent); -bnxt_set_max_func_cp_rings (bp, max_cp_rings - avail_msix); edev->flags |=BNXT_EN_FLAG_MSIX_REQUESTED ;return avail_msix; }
@@ -178,7 +177,6 @@ static int bnxt_free_msix_vecs(struct bn {struct net_device *dev = edev->net;struct bnxt *bp =netdev_priv (dev); -int max_cp_rings, msix_requested;ASSERT_RTNL ();if (ulp_id !=BNXT_ROCE_ULP )
@@ -187,9 +185,6 @@ static int bnxt_free_msix_vecs(struct bnif (!(edev->flags &BNXT_EN_FLAG_MSIX_REQUESTED ))return 0 ; - max_cp_rings =bnxt_get_max_func_cp_rings (bp); - msix_requested = edev->ulp_tbl[ulp_id].msix_requested; -bnxt_set_max_func_cp_rings (bp, max_cp_rings + msix_requested); edev->ulp_tbl[ulp_id].msix_requested =0 ; edev->flags &= ~BNXT_EN_FLAG_MSIX_REQUESTED ;if (netif_running (dev)) {
Greg Kroah-Hartman
maintainerPATCH 16/197 net/sched: act_pedit: fix dump of extended layered op
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Davide Caratti <[email protected]>
[ Upstream commit 85eb9af182243ce9a8b72410d5321c440ac5f8d7 ]
in the (rare) case of failure in nla_nest_start(), missing NULL checks in tcf_pedit_key_ex_dump() can make the following command
# tc action add action pedit ex munge ip ttl set 64
dereference a NULL pointer:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000000 PGD 800000007d1cd067 P4D 800000007d1cd067 PUD 7acd3067 PMD 0 Oops: 0002 [#1] SMP PTI CPU: 0 PID: 3336 Comm: tc Tainted: G E 4.18.0.pedit+ #425 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 RIP: 0010:tcf_pedit_dump+0x19d/0x358 [act_pedit] Code: be 02 00 00 00 48 89 df 66 89 44 24 20 e8 9b b1 fd e0 85 c0 75 46 8b 83 c8 00 00 00 49 83 c5 08 48 03 83 d0 00 00 00 4d 39 f5 <66> 89 04 25 00 00 00 00 0f 84 81 01 00 00 41 8b 45 00 48 8d 4c 24 RSP: 0018:ffffb5d4004478a8 EFLAGS: 00010246 RAX: ffff8880fcda2070 RBX: ffff8880fadd2900 RCX: 0000000000000000 RDX: 0000000000000002 RSI: ffffb5d4004478ca RDI: ffff8880fcda206e RBP: ffff8880fb9cb900 R08: 0000000000000008 R09: ffff8880fcda206e R10: ffff8880fadd2900 R11: 0000000000000000 R12: ffff8880fd26cf40 R13: ffff8880fc957430 R14: ffff8880fc957430 R15: ffff8880fb9cb988 FS: 00007f75a537a740(0000) GS:ffff8880fda00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000007a2fa005 CR4: 00000000001606f0 Call Trace: ? __nla_reserve+0x38/0x50 tcf_action_dump_1+0xd2/0x130 tcf_action_dump+0x6a/0xf0 tca_get_fill.constprop.31+0xa3/0x120 tcf_action_add+0xd1/0x170 tc_ctl_action+0x137/0x150 rtnetlink_rcv_msg+0x263/0x2d0 ? _cond_resched+0x15/0x40 ? rtnl_calcit.isra.30+0x110/0x110 netlink_rcv_skb+0x4d/0x130 netlink_unicast+0x1a3/0x250 netlink_sendmsg+0x2ae/0x3a0 sock_sendmsg+0x36/0x40 ___sys_sendmsg+0x26f/0x2d0 ? do_wp_page+0x8e/0x5f0 ? handle_pte_fault+0x6c3/0xf50 ? __handle_mm_fault+0x38e/0x520 ? __sys_sendmsg+0x5e/0xa0 __sys_sendmsg+0x5e/0xa0 do_syscall_64+0x5b/0x180 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x7f75a4583ba0 Code: c3 48 8b 05 f2 62 2c 00 f7 db 64 89 18 48 83 cb ff eb dd 0f 1f 80 00 00 00 00 83 3d fd c3 2c 00 00 75 10 b8 2e 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 ae cc 00 00 48 89 04 24 RSP: 002b:00007fff60ee7418 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 00007fff60ee7540 RCX: 00007f75a4583ba0 RDX: 0000000000000000 RSI: 00007fff60ee7490 RDI: 0000000000000003 RBP: 000000005b842d3e R08: 0000000000000002 R09: 0000000000000000 R10: 00007fff60ee6ea0 R11: 0000000000000246 R12: 0000000000000000 R13: 00007fff60ee7554 R14: 0000000000000001 R15: 000000000066c100 Modules linked in: act_pedit(E) ip6table_filter ip6_tables iptable_filter binfmt_misc crct10dif_pclmul ext4 crc32_pclmul mbcache ghash_clmulni_intel jbd2 pcbc snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep snd_seq snd_seq_device snd_pcm aesni_intel crypto_simd snd_timer cryptd glue_helper snd joydev pcspkr soundcore virtio_balloon i2c_piix4 nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables xfs libcrc32c ata_generic pata_acpi virtio_net net_failover virtio_blk virtio_console failover qxl crc32c_intel drm_kms_helper syscopyarea serio_raw sysfillrect sysimgblt fb_sys_fops ttm drm ata_piix virtio_pci libata virtio_ring i2c_core virtio floppy dm_mirror dm_region_hash dm_log dm_mod [last unloaded: act_pedit] CR2: 0000000000000000
Like it's done for other TC actions, give up dumping pedit rules and return an error if nla_nest_start() returns NULL.
--- net/sched/act_pedit.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-)
net/sched/act_pedit.c +13 -5
@@ -109,16 +109,18 @@ static int tcf_pedit_key_ex_dump(struct {struct nlattr *keys_start =nla_nest_start (skb,TCA_PEDIT_KEYS_EX ); +if (!keys_start) + goto nla_failure;for (; n >0 ; n--) {struct nlattr *key_start; key_start =nla_nest_start (skb,TCA_PEDIT_KEY_EX ); +if (!key_start) + goto nla_failure;if (nla_put_u16 (skb,TCA_PEDIT_KEY_EX_HTYPE , keys_ex->htype) || -nla_put_u16 (skb,TCA_PEDIT_KEY_EX_CMD , keys_ex->cmd)) { -nlmsg_trim (skb, keys_start); -return -EINVAL ; - } +nla_put_u16 (skb,TCA_PEDIT_KEY_EX_CMD , keys_ex->cmd)) + goto nla_failure;nla_nest_end (skb, key_start);
@@ -128,6 +130,9 @@ static int tcf_pedit_key_ex_dump(structnla_nest_end (skb, keys_start);return 0 ; +nla_failure: +nla_nest_cancel (skb, keys_start); +return -EINVAL ; } static int tcf_pedit_init(struct net *net, struct nlattr *nla,
@@ -395,7 +400,10 @@ static int tcf_pedit_dump(struct sk_buff opt->bindcnt = p->tcf_bindcnt - bind;if (p->tcfp_keys_ex) { -tcf_pedit_key_ex_dump (skb, p->tcfp_keys_ex, p->tcfp_nkeys); +if (tcf_pedit_key_ex_dump (skb, + p->tcfp_keys_ex, + p->tcfp_nkeys)) + goto nla_put_failure;if (nla_put (skb,TCA_PEDIT_PARMS_EX , s, opt)) goto nla_put_failure;
Greg Kroah-Hartman
maintainerPATCH 17/197 tipc: fix a missing rhashtable_walk_exit()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Cong Wang <[email protected]>
[ Upstream commit bd583fe30427500a2d0abe25724025b1cb5e2636 ]
rhashtable_walk_exit() must be paired with rhashtable_walk_enter().
--- net/tipc/socket.c | 2 ++ 1 file changed, 2 insertions(+)
net/tipc/socket.c +2 -0
@@ -2675,6 +2675,8 @@ void tipc_sk_reinit(struct net *net)rhashtable_walk_stop (&iter); }while (tsk ==ERR_PTR (-EAGAIN )); + +rhashtable_walk_exit (&iter); }static struct tipc_sock *tipc_sk_lookup (struct net *net,u32 portid)
Greg Kroah-Hartman
maintainerPATCH 18/197 hv_netvsc: Fix a deadlock by getting rtnl lock earlier in netvsc_probe()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dexuan Cui <[email protected]>
[ Upstream commit e04e7a7bbd4bbabef4e1a58367e5fc9b2edc3b10 ]
This patch fixes the race between netvsc_probe() and rndis_set_subchannel(), which can cause a deadlock.
These are the related 3 paths which show the deadlock:
path #1:
Workqueue: hv_vmbus_con vmbus_onmessage_work [hv_vmbus]
Call Trace:
schedule
schedule_preempt_disabled
__mutex_lock
__device_attach
bus_probe_device
device_add
vmbus_device_register
vmbus_onoffer
vmbus_onmessage_work
process_one_work
worker_thread
kthread
ret_from_forkpath #2:
schedule
schedule_preempt_disabled
__mutex_lock
netvsc_probe
vmbus_probe
really_probe
__driver_attach
bus_for_each_dev
driver_attach_async
async_run_entry_fn
process_one_work
worker_thread
kthread
ret_from_fork
path #3:
Workqueue: events netvsc_subchan_work [hv_netvsc]
Call Trace:
schedule
rndis_set_subchannel
netvsc_subchan_work
process_one_work
worker_thread
kthread
ret_from_forkBefore path #1 finishes, path #2 can start to run, because just before the "bus_probe_device(dev);" in device_add() in path #1, there is a line "object_uevent(&dev->kobj, KOBJ_ADD);", so systemd-udevd can immediately try to load hv_netvsc and hence path #2 can start to run.
Next, path #2 offloads the subchannal's initialization to a workqueue, i.e. path #3, so we can end up in a deadlock situation like this:
Path #2 gets the device lock, and is trying to get the rtnl lock; Path #3 gets the rtnl lock and is waiting for all the subchannel messages to be processed;
Path #1 is trying to get the device lock, but since #2 is not releasing the device lock, path #1 has to sleep; since the VMBus messages are processed one by one, this means the sub-channel messages can't be procedded, so #3 has to sleep with the rtnl lock held, and finally #2 has to sleep... Now all the 3 paths are sleeping and we hit the deadlock.
With the patch, we can make sure #2 gets both the device lock and the rtnl lock together, gets its job done, and releases the locks, so #1 and #3 will not be blocked for ever.
--- drivers/net/hyperv/netvsc_drv.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)
drivers/net/hyperv/netvsc_drv.c +10 -1
@@ -2101,6 +2101,16 @@ static int netvsc_probe(struct hv_devicememcpy (net->dev_addr, device_info.mac_adr,ETH_ALEN ); +/* We must get rtnl lock before scheduling nvdev->subchan_work, +* otherwise netvsc_subchan_work() can get rtnl lock first and wait +* all subchannels to show up, but that may not happen because +* netvsc_probe() can't get rtnl lock and as a result vmbus_onoffer() +* -> ... -> device_add() -> ... -> __device_attach() can't get +* the device lock, so all the subchannels can't be processed -- +* finally netvsc_subchan_work() hangs for ever. +*/ +rtnl_lock (); +if (nvdev->num_chn >1 )schedule_work (&nvdev->subchan_work);
@@ -2119,7 +2129,6 @@ static int netvsc_probe(struct hv_deviceelse net->max_mtu =ETH_DATA_LEN ; -rtnl_lock (); ret =register_netdevice (net);if (ret !=0 ) {pr_err ("Unable to register netdev.\n" );
Greg Kroah-Hartman
maintainerPATCH 19/197 tipc: fix the big/little endian issue in tipc_dest
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Haiqing Bai <[email protected]>
[ Upstream commit 30935198b7d0be12b1c45c328b66a7fdefb16256 ]
In function tipc_dest_push, the 32bit variables 'node' and 'port' are stored separately in uppper and lower part of 64bit 'value'. Then this value is assigned to dst->value which is a union like:
union
{
struct {
u32 port;
u32 node; };
u64 value;
}
This works on little-endian machines like x86 but fails on big-endian machines.
The fix remove the 'value' stack parameter and even the 'value' member of the union in tipc_dest, assign the 'node' and 'port' member directly with the input parameter to avoid the endian issue.
--- net/tipc/name_table.c | 10 ++++------ net/tipc/name_table.h | 9 ++------- 2 files changed, 6 insertions(+), 13 deletions(-)
net/tipc/name_table.c +4 -6
@@ -980,20 +980,17 @@ int tipc_nl_name_table_dump(struct sk_bustruct tipc_dest *tipc_dest_find (struct list_head *l,u32 node,u32 port) { -u64 value = (u64 )node <<32 | port;struct tipc_dest *dst;list_for_each_entry (dst, l, list) { -if (dst->value != value) -continue ; -return dst; +if (dst->node == node && dst->port == port) +return dst; }return NULL ; }bool tipc_dest_push (struct list_head *l,u32 node,u32 port) { -u64 value = (u64 )node <<32 | port;struct tipc_dest *dst;if (tipc_dest_find (l, node, port))
@@ -1002,7 +999,8 @@ bool tipc_dest_push(struct list_head *l, dst =kmalloc (sizeof (*dst),GFP_ATOMIC );if (unlikely (!dst))return false; - dst->value = value; + dst->node = node; + dst->port = port;list_add (&dst->list, l);return true; }
net/tipc/name_table.h +2 -7
@@ -133,13 +133,8 @@ void tipc_nametbl_stop(struct net *net);struct tipc_dest {struct list_head list; -union { -struct { -u32 port; -u32 node; - }; -u64 value; - }; +u32 port; +u32 node; };struct tipc_dest *tipc_dest_find (struct list_head *l,u32 node,u32 port);
Greg Kroah-Hartman
maintainerPATCH 20/197 sctp: remove useless start_fail from sctp_ht_iter in proc
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xin Long <[email protected]>
[ Upstream commit 834539e69a5fe2aab33cc777ccfd4a4fcc5b9770 ]
After changing rhashtable_walk_start to return void, start_fail would never be set other value than 0, and the checking for start_fail is pointless, so remove it.
--- net/sctp/proc.c | 4 ---- 1 file changed, 4 deletions(-)
net/sctp/proc.c +0 -4
@@ -215,7 +215,6 @@ static const struct seq_operations sctp_struct sctp_ht_iter {struct seq_net_private p;struct rhashtable_iter hti; -int start_fail; };static void *sctp_transport_seq_start (struct seq_file *seq,loff_t *pos)
@@ -224,7 +223,6 @@ static void *sctp_transport_seq_start(stsctp_transport_walk_start (&iter->hti); - iter->start_fail =0 ;return sctp_transport_get_idx (seq_file_net (seq), &iter->hti, *pos); }
@@ -232,8 +230,6 @@ static void sctp_transport_seq_stop(stru {struct sctp_ht_iter *iter = seq->private; -if (iter->start_fail) -return ;sctp_transport_walk_stop (&iter->hti); }
Greg Kroah-Hartman
maintainerPATCH 21/197 erspan: set erspan_ver to 1 by default when adding an erspan dev
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xin Long <[email protected]>
[ Upstream commit 84581bdae9587023cea1d139523f0ef0f28bd88d ]
After erspan_ver is introudced, if erspan_ver is not set in iproute, its value will be left 0 by default. Since Commit 02f99df1875c ("erspan: fix invalid erspan version."), it has broken the traffic due to the version check in erspan_xmit if users are not aware of 'erspan_ver' param, like using an old version of iproute.
To fix this compatibility problem, it sets erspan_ver to 1 by default when adding an erspan dev in erspan_setup. Note that we can't do it in ipgre_netlink_parms, as this function is also used by ipgre_changelink.
--- net/ipv4/ip_gre.c | 3 +++ net/ipv6/ip6_gre.c | 1 + 2 files changed, 4 insertions(+)
net/ipv4/ip_gre.c +3 -0
@@ -1511,11 +1511,14 @@ nla_put_failure:static void erspan_setup (struct net_device *dev) { +struct ip_tunnel *t =netdev_priv (dev); +ether_setup (dev); dev->netdev_ops = &erspan_netdev_ops; dev->priv_flags &= ~IFF_TX_SKB_SHARING ; dev->priv_flags |=IFF_LIVE_ADDR_CHANGE ;ip_tunnel_setup (dev, erspan_net_id); + t->erspan_ver =1 ; }static const struct nla_policy ipgre_policy[IFLA_GRE_MAX +1 ] = {
net/ipv6/ip6_gre.c +1 -0
@@ -1776,6 +1776,7 @@ static void ip6gre_netlink_parms(structif (data[IFLA_GRE_COLLECT_METADATA ]) parms->collect_md = true; + parms->erspan_ver =1 ;if (data[IFLA_GRE_ERSPAN_VER ]) parms->erspan_ver =nla_get_u8 (data[IFLA_GRE_ERSPAN_VER ]);
Greg Kroah-Hartman
maintainerPATCH 22/197 net: macb: Fix regression breaking non-MDIO fixed-link PHYs
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ahmad Fatoum <[email protected]>
[ Upstream commit ab5f11055fdf8dfc3ddbd89e8e3cc550de41d1d3 ]
commit 739de9a1563a ("net: macb: Reorganize macb_mii bringup") broke initializing macb on the EVB-KSZ9477 eval board.
There, of_mdiobus_register was called even for the fixed-link representing the RGMII-link to the switch with the result that the driver attempts to enumerate PHYs on a non-existent MDIO bus:
libphy: MACB_mii_bus: probed
mdio_bus f0028000.ethernet-ffffffff: fixed-link has invalid PHY address mdio_bus f0028000.ethernet-ffffffff: scan phy fixed-link at address 0 [snip]
mdio_bus f0028000.ethernet-ffffffff: scan phy fixed-link at address 31
The "MDIO" bus registration succeeds regardless, having claimed the reset GPIO, and calling of_phy_register_fixed_link later on fails because it tries to claim the same GPIO:
macb f0028000.ethernet: broken fixed-link specification
Fix this by registering the fixed-link before calling mdiobus_register.
--- drivers/net/ethernet/cadence/macb_main.c | 27 +++++++++++++++++---------- 1 file changed, 17 insertions(+), 10 deletions(-)
drivers/net/ethernet/cadence/macb_main.c +17 -10
@@ -481,11 +481,6 @@ static int macb_mii_probe(struct net_devif (np) {if (of_phy_is_fixed_link (np)) { -if (of_phy_register_fixed_link (np) <0 ) { -dev_err (&bp->pdev->dev, -"broken fixed-link specification\n" ); -return -ENODEV ; - } bp->phy_node =of_node_get (np); }else { bp->phy_node =of_parse_phandle (np,"phy-handle" ,0 );
@@ -568,7 +563,7 @@ static int macb_mii_init(struct macb *bp {struct macb_platform_data *pdata;struct device_node *np; -int err; +int err = -ENXIO ;/* Enable management port */ macb_writel (bp,NCR ,MACB_BIT (MPE ));
@@ -591,12 +586,23 @@ static int macb_mii_init(struct macb *bpdev_set_drvdata (&bp->dev->dev, bp->mii_bus); np = bp->pdev->dev.of_node; -if (pdata) - bp->mii_bus->phy_mask = pdata->phy_mask; +if (np &&of_phy_is_fixed_link (np)) { +if (of_phy_register_fixed_link (np) <0 ) { +dev_err (&bp->pdev->dev, +"broken fixed-link specification %pOF\n" , np); + goto err_out_free_mdiobus; + } + + err =mdiobus_register (bp->mii_bus); + }else { +if (pdata) + bp->mii_bus->phy_mask = pdata->phy_mask; + + err =of_mdiobus_register (bp->mii_bus, np); + } - err =of_mdiobus_register (bp->mii_bus, np);if (err) - goto err_out_free_mdiobus; + goto err_out_free_fixed_link; err =macb_mii_probe (bp->dev);if (err)
@@ -606,6 +612,7 @@ static int macb_mii_init(struct macb *bp err_out_unregister_bus:mdiobus_unregister (bp->mii_bus); +err_out_free_fixed_link:if (np &&of_phy_is_fixed_link (np))of_phy_deregister_fixed_link (np); err_out_free_mdiobus:
Greg Kroah-Hartman
maintainerPATCH 23/197 ipv6: dont get lwtstate twice in ip6_rt_copy_init()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexey Kodanev <[email protected]>
[ Upstream commit 93bbadd6e0a2a58e49d265b9b1aa58e621b60a26 ]
Commit 80f1a0f4e0cd ("net/ipv6: Put lwtstate when destroying fib6_info") partially fixed the kmemleak [1], lwtstate can be copied from fib6_info, with ip6_rt_copy_init(), and it should be done only once there.
rt->dst.lwtstate is set by ip6_rt_init_dst(), at the start of the function ip6_rt_copy_init(), so there is no need to get it again at the end.
With this patch, lwtstate also isn't copied from RTF_REJECT routes.
[1]:
unreferenced object 0xffff880b6aaa14e0 (size 64):
comm "ip", pid 10577, jiffies 4295149341 (age 1273.903s) hex dump (first 32 bytes):
01 00 04 00 04 00 00 00 10 00 00 00 00 00 00 00 ................
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<0000000018664623>] lwtunnel_build_state+0x1bc/0x420
[<00000000b73aa29a>] ip6_route_info_create+0x9f7/0x1fd0
[<00000000ee2c5d1f>] ip6_route_add+0x14/0x70
[<000000008537b55c>] inet6_rtm_newroute+0xd9/0xe0
[<000000002acc50f5>] rtnetlink_rcv_msg+0x66f/0x8e0
[<000000008d9cd381>] netlink_rcv_skb+0x268/0x3b0
[<000000004c893c76>] netlink_unicast+0x417/0x5a0
[<00000000f2ab1afb>] netlink_sendmsg+0x70b/0xc30
[<00000000890ff0aa>] sock_sendmsg+0xb1/0xf0
[<00000000a2e7b66f>] ___sys_sendmsg+0x659/0x950
[<000000001e7426c8>] __sys_sendmsg+0xde/0x170
[<00000000fe411443>] do_syscall_64+0x9f/0x4a0
[<000000001be7b28b>] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[<000000006d21f353>] 0xffffffffffffffff--- net/ipv6/route.c | 1 - 1 file changed, 1 deletion(-)
net/ipv6/route.c +0 -1
@@ -996,7 +996,6 @@ static void ip6_rt_copy_init(struct rt6_ rt->rt6i_src = ort->fib6_src;#endif rt->rt6i_prefsrc = ort->fib6_prefsrc; - rt->dst.lwtstate =lwtstate_get (ort->fib6_nh.nh_lwtstate); } static struct fib6_node* fib6_backtrack(struct fib6_node *fn,
Greg Kroah-Hartman
maintainerPATCH 24/197 net/ipv6: init ip6 anycast rt->dst.input as ip6_input
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hangbin Liu <[email protected]>
[ Upstream commit d23c4b6336ef30898dcdff351f21e633e7a64930 ]
Commit 6edb3c96a5f02 ("net/ipv6: Defer initialization of dst to data path") forgot to handle anycast route and init anycast rt->dst.input to ip6_forward. Fix it by setting anycast rt->dst.input back to ip6_input.
--- net/ipv6/route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/route.c +1 -1
@@ -956,7 +956,7 @@ static void ip6_rt_init_dst(struct rt6_i rt->dst.error =0 ; rt->dst.output = ip6_output; -if (ort->fib6_type ==RTN_LOCAL ) { +if (ort->fib6_type ==RTN_LOCAL || ort->fib6_type ==RTN_ANYCAST ) { rt->dst.input = ip6_input; }else if (ipv6_addr_type (&ort->fib6_dst.addr) &IPV6_ADDR_MULTICAST ) { rt->dst.input = ip6_mc_input;
Greg Kroah-Hartman
maintainerPATCH 25/197 net/ipv6: Only update MTU metric if it set
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Ahern <[email protected]>
[ Upstream commit 15a81b418e22a9aa4a0504471fdcb0f4ebf69b96 ]
Jan reported a regression after an update to 4.18.5. In this case ipv6 default route is setup by systemd-networkd based on data from an RA. The RA contains an MTU of 1492 which is used when the route is first inserted but then systemd-networkd pushes down updates to the default route without the mtu set.
Prior to the change to fib6_info, metrics such as MTU were held in the dst_entry and rt6i_pmtu in rt6_info contained an update to the mtu if any. ip6_mtu would look at rt6i_pmtu first and use it if set. If not, the value from the metrics is used if it is set and finally falling back to the idev value.
After the fib6_info change metrics are contained in the fib6_info struct and there is no equivalent to rt6i_pmtu. To maintain consistency with the old behavior the new code should only reset the MTU in the metrics if the route update has it set.
--- net/ipv6/ip6_fib.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/ipv6/ip6_fib.c +4 -1
@@ -987,7 +987,10 @@ static int fib6_add_rt2node(struct fib6_fib6_clean_expires (iter); else fib6_set_expires(iter, rt->expires); -fib6_metric_set (iter,RTAX_MTU , rt->fib6_pmtu); + +if (rt->fib6_pmtu) +fib6_metric_set (iter,RTAX_MTU , + rt->fib6_pmtu);return -EEXIST ; }/* If we have the same destination and the same metric,
Greg Kroah-Hartman
maintainerPATCH 26/197 net/ipv6: Put lwtstate when destroying fib6_info
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Ahern <[email protected]>
[ Upstream commit 80f1a0f4e0cd4bfc8a74fc1c39843a6e7b206b95 ]
Prior to the introduction of fib6_info lwtstate was managed by the dst code. With fib6_info releasing lwtstate needs to be done when the struct is freed.
--- net/ipv6/ip6_fib.c | 2 ++ 1 file changed, 2 insertions(+)
net/ipv6/ip6_fib.c +2 -0
@@ -198,6 +198,8 @@ void fib6_info_destroy_rcu(struct rcu_he } } +lwtstate_put (f6i->fib6_nh.nh_lwtstate); +if (f6i->fib6_nh.nh_dev)dev_put (f6i->fib6_nh.nh_dev);
Greg Kroah-Hartman
maintainerPATCH 27/197 net/mlx5: Fix SQ offset in QPs with small RQ
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tariq Toukan <[email protected]>
[ Upstream commit 639505d4397b8c654a8e2616f9cb70ece40c83f9 ]
Correct the formula for calculating the RQ page remainder, which should be in byte granularity. The result will be non-zero only for RQs smaller than PAGE_SIZE, as an RQ size is a power of 2.
Divide this by the SQ stride (MLX5_SEND_WQE_BB) to get the SQ offset in strides granularity.
--- drivers/net/ethernet/mellanox/mlx5/core/wq.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/wq.c +3 -2
@@ -139,14 +139,15 @@ int mlx5_wq_qp_create(struct mlx5_core_d struct mlx5_wq_ctrl *wq_ctrl) {u32 sq_strides_offset; +u32 rq_pg_remainder;int err;mlx5_fill_fbc (MLX5_GET (qpc, qpc, log_rq_stride) +4 ,MLX5_GET (qpc, qpc, log_rq_size), &wq->rq.fbc); - sq_strides_offset = - ((wq->rq.fbc.frag_sz_m1 +1 ) %PAGE_SIZE ) /MLX5_SEND_WQE_BB ; + rq_pg_remainder =mlx5_wq_cyc_get_byte_size (&wq->rq) %PAGE_SIZE ; + sq_strides_offset = rq_pg_remainder /MLX5_SEND_WQE_BB ; mlx5_fill_fbc_offset(ilog2(MLX5_SEND_WQE_BB), MLX5_GET(qpc, qpc, log_sq_size),
Greg Kroah-Hartman
maintainerPATCH 28/197 r8169: set RxConfig after tx/rx is enabled for RTL8169sb/8110sb devices
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Azat Khuzhin <[email protected]>
[ Upstream commit 05212ba8132b42047ab5d63d759c6f9c28e7eab5 ]
I have two Ethernet adapters:
r8169 0000:03:01.0 eth0: RTL8169sb/8110sb, 00:14:d1:14:2d:49, XID 10000000, IRQ 18 r8169 0000:01:00.0 eth0: RTL8168e/8111e, 64:66:b3:11:14:5d, XID 2c200000, IRQ 30
And after upgrading from linux 4.15 [1] to linux 4.18+ [2] RTL8169sb failed to receive any packets. tcpdump shows a lot of checksum mismatch.
[1]: a0f79386a4968b4925da6db2d1daffd0605a4402
[2]: 0519359784328bfa92bf0931bf0cff3b58c16932 (4.19 merge window opened)
I started bisecting and the found that [3] breaks it. According to [4]:
"For 8110S, 8110SB, and 8110SC series, the initial value of RxConfig needs to be set after the tx/rx is enabled."
So I moved rtl_init_rxcfg() after enabling tx/rs and now my adapter works (RTL8168e works too).
[3]: 3559d81e76bfe3803e89f2e04cf6ef7ab4f3aace
[4]: e542a2269f232d61270ceddd42b73a4348dee2bb ("r8169: adjust the RxConfig
settings.")
Also drop "rx" from rtl_set_rx_tx_config_registers(), since it does nothing with it already.
rtl_hw_start_8169")
--- drivers/net/ethernet/realtek/r8169.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/ethernet/realtek/r8169.c +4 -2
@@ -5039,7 +5039,7 @@ static void rtl8169_hw_reset(struct rtl8rtl_hw_reset (tp); } -static void rtl_set_rx_tx_config_registers (struct rtl8169_private *tp) +static void rtl_set_tx_config_registers (struct rtl8169_private *tp) {/* Set DMA burst size and Interframe Gap Time */ RTL_W32(tp, TxConfig, (TX_DMA_BURST << TxDMAShift) |
@@ -5150,12 +5150,14 @@ static void rtl_hw_start(struct rtl8169rtl_set_rx_max_size (tp);rtl_set_rx_tx_desc_registers (tp); -rtl_set_rx_tx_config_registers (tp); +rtl_set_tx_config_registers (tp);RTL_W8 (tp, Cfg9346, Cfg9346_Lock);/* Initially a 10 us delay. Turned it into a PCI commit. - FR */ RTL_R8 (tp, IntrMask);RTL_W8 (tp, ChipCmd, CmdTxEnb | CmdRxEnb); +rtl_init_rxcfg (tp); +rtl_set_rx_mode (tp->dev);/* no early-rx interrupts */ RTL_W16 (tp, MultiIntr,RTL_R16 (tp, MultiIntr) &0xf000 );
Greg Kroah-Hartman
maintainerPATCH 29/197 Revert "net: stmmac: Do not keep rearming the coalesce timer in stmmac_xmit"
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jerome Brunet <[email protected]>
[ Upstream commit e5133f2f1261f8ab412e7fc5e3694c9f84328f89 ]
This reverts commit 4ae0169fd1b3c792b66be58995b7e6b629919ecf.
This change in the handling of the coalesce timer is causing regression on (at least) amlogic platforms.
Network will break down very quickly (a few seconds) after starting a download. This can easily be reproduced using iperf3 for example.
The problem has been reported on the S805, S905, S912 and A113 SoCs (Realtek and Micrel PHYs) and it is likely impacting all Amlogics platforms using Gbit ethernet
No problem was seen with the platform using 10/100 only PHYs (GXL internal)
Reverting change brings things back to normal and allows to use network again until we better understand the problem with the coalesce timer.
--- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 +---- 2 files changed, 1 insertion(+), 5 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac.h +0 -1
@@ -112,7 +112,6 @@ struct stmmac_priv {u32 tx_count_frames;u32 tx_coal_frames;u32 tx_coal_timer; -bool tx_timer_armed;int tx_coalesce;int hwts_tx_en;
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +1 -4
@@ -3126,16 +3126,13 @@ static netdev_tx_t stmmac_xmit(struct sk* element in case of no SG. */ priv->tx_count_frames += nfrags +1 ; -if (likely (priv->tx_coal_frames > priv->tx_count_frames) && - !priv->tx_timer_armed) { +if (likely (priv->tx_coal_frames > priv->tx_count_frames)) {mod_timer (&priv->txtimer,STMMAC_COAL_TIMER (priv->tx_coal_timer)); - priv->tx_timer_armed = true; }else { priv->tx_count_frames =0 ;stmmac_set_tx_ic (priv, desc); priv->xstats.tx_set_ic_bit++; - priv->tx_timer_armed = false; }skb_tx_timestamp (skb);
Greg Kroah-Hartman
maintainerPATCH 30/197 ip6_vti: fix creating fallback tunnel device for vti6
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Haishuang Yan <[email protected]>
[ Upstream commit e2948e5af8eeb6c945000772b7613b0323a0a203 ]
When set fb_tunnels_only_for_init_net to 1, don't create fallback tunnel device for vti6 when a new namespace is created.
Tested:
[root@builder2 ~]# modprobe ip6_tunnel
[root@builder2 ~]# modprobe ip6_vti
[root@builder2 ~]# echo 1 > /proc/sys/net/core/fb_tunnels_only_for_init_net [root@builder2 ~]# unshare -n
[root@builder2 ~]# ip link
1: lo: <LOOPBACK> mtu 65536 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
--- net/ipv6/ip6_vti.c | 2 ++ 1 file changed, 2 insertions(+)
net/ipv6/ip6_vti.c +2 -0
@@ -1114,6 +1114,8 @@ static int __net_init vti6_init_net(stru ip6n->tnls[0 ] = ip6n->tnls_wc; ip6n->tnls[1 ] = ip6n->tnls_r_l; +if (!net_has_fallback_tunnels (net)) +return 0 ; err = -ENOMEM ; ip6n->fb_tnl_dev =alloc_netdev (sizeof (struct ip6_tnl ),"ip6_vti0" ,NET_NAME_UNKNOWN , vti6_dev_setup);
Greg Kroah-Hartman
maintainerPATCH 31/197 ip6_vti: fix a null pointer deference when destroy vti6 tunnel
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Haishuang Yan <[email protected]>
[ Upstream commit 9c86336c15db1c48cbaddff56caf2be0a930e991 ]
If load ip6_vti module and create a network namespace when set fb_tunnels_only_for_init_net to 1, then exit the namespace will cause following crash:
[ 6601.677036] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 [ 6601.679057] PGD 8000000425eca067 P4D 8000000425eca067 PUD 424292067 PMD 0 [ 6601.680483] Oops: 0000 [#1] SMP PTI [ 6601.681223] CPU: 7 PID: 93 Comm: kworker/u16:1 Kdump: loaded Tainted: G E 4.18.0+ #3 [ 6601.683153] Hardware name: Fedora Project OpenStack Nova, BIOS seabios-1.7.5-11.el7 04/01/2014 [ 6601.684919] Workqueue: netns cleanup_net [ 6601.685742] RIP: 0010:vti6_exit_batch_net+0x87/0xd0 [ip6_vti] [ 6601.686932] Code: 7b 08 48 89 e6 e8 b9 ea d3 dd 48 8b 1b 48 85 db 75 ec 48 83 c5 08 48 81 fd 00 01 00 00 75 d5 49 8b 84 24 08 01 00 00 48 89 e6 <48> 8b 78 08 e8 90 ea d3 dd 49 8b 45 28 49 39 c6 4c 8d 68 d8 75 a1 [ 6601.690735] RSP: 0018:ffffa897c2737de0 EFLAGS: 00010246 [ 6601.691846] RAX: 0000000000000000 RBX: 0000000000000000 RCX: dead000000000200 [ 6601.693324] RDX: 0000000000000015 RSI: ffffa897c2737de0 RDI: ffffffff9f2ea9e0 [ 6601.694824] RBP: 0000000000000100 R08: 0000000000000000 R09: 0000000000000000 [ 6601.696314] R10: 0000000000000001 R11: 0000000000000000 R12: ffff8dc323c07e00 [ 6601.697812] R13: ffff8dc324a63100 R14: ffffa897c2737e30 R15: ffffa897c2737e30 [ 6601.699345] FS: 0000000000000000(0000) GS:ffff8dc33fdc0000(0000) knlGS:0000000000000000 [ 6601.701068] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 6601.702282] CR2: 0000000000000008 CR3: 0000000424966002 CR4: 00000000001606e0 [ 6601.703791] Call Trace: [ 6601.704329] cleanup_net+0x1b4/0x2c0 [ 6601.705268] process_one_work+0x16c/0x370 [ 6601.706145] worker_thread+0x49/0x3e0 [ 6601.706942] kthread+0xf8/0x130 [ 6601.707626] ? rescuer_thread+0x340/0x340 [ 6601.708476] ? kthread_bind+0x10/0x10 [ 6601.709266] ret_from_fork+0x35/0x40
Reproduce:
modprobe ip6_vti
echo 1 > /proc/sys/net/core/fb_tunnels_only_for_init_net unshare -n
exit
This because ip6n->tnls_wc[0] point to fallback device in default, but in non-default namespace, ip6n->tnls_wc[0] will be NULL, so add the NULL check comparatively.
--- net/ipv6/ip6_vti.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/ipv6/ip6_vti.c +2 -1
@@ -1102,7 +1102,8 @@ static void __net_exit vti6_destroy_tunn } t =rtnl_dereference (ip6n->tnls_wc[0 ]); -unregister_netdevice_queue (t->dev, list); +if (t) +unregister_netdevice_queue (t->dev, list); }static int __net_init vti6_init_net (struct net *net)
Greg Kroah-Hartman
maintainerPATCH 32/197 nfp: wait for posted reconfigs when disabling the device
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jakub Kicinski <[email protected]>
[ Upstream commit 9ad716b95fd6c6be46a4f2d5936e514b5bcd744d ]
To avoid leaking a running timer we need to wait for the posted reconfigs after netdev is unregistered. In common case the process of deinitializing the device will perform synchronous reconfigs which wait for posted requests, but especially with VXLAN ports being actively added and removed there can be a race condition leaving a timer running after adapter structure is freed leading to a crash.
Add an explicit flush after deregistering and for a good measure a warning to check if timer is running just before structures are freed.
--- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 48 +++++++++++++------- 1 file changed, 33 insertions(+), 15 deletions(-)
drivers/net/ethernet/netronome/nfp/nfp_net_common.c +33 -15
@@ -227,29 +227,16 @@ done:spin_unlock_bh (&nn->reconfig_lock); } -/** -* nfp_net_reconfig() - Reconfigure the firmware -* @nn: NFP Net device to reconfigure -* @update: The value for the update field in the BAR config -* -* Write the update word to the BAR and ping the reconfig queue. The -* poll until the firmware has acknowledged the update by zeroing the -* update word. -* -* Return: Negative errno on error, 0 on success -*/ -int nfp_net_reconfig (struct nfp_net *nn,u32 update) +static void nfp_net_reconfig_sync_enter (struct nfp_net *nn) {bool cancelled_timer = false;u32 pre_posted_requests; -int ret;spin_lock_bh (&nn->reconfig_lock); nn->reconfig_sync_present = true;if (nn->reconfig_timer_active) { -del_timer (&nn->reconfig_timer); nn->reconfig_timer_active = false; cancelled_timer = true; }
@@ -258,14 +245,43 @@ int nfp_net_reconfig(struct nfp_net *nn,spin_unlock_bh (&nn->reconfig_lock); -if (cancelled_timer) +if (cancelled_timer) { +del_timer_sync (&nn->reconfig_timer);nfp_net_reconfig_wait (nn, nn->reconfig_timer.expires); + }/* Run the posted reconfigs which were issued before we started */ if (pre_posted_requests) {nfp_net_reconfig_start (nn, pre_posted_requests);nfp_net_reconfig_wait (nn, jiffies +HZ *NFP_NET_POLL_TIMEOUT ); } +} + +static void nfp_net_reconfig_wait_posted (struct nfp_net *nn) +{ +nfp_net_reconfig_sync_enter (nn); + +spin_lock_bh (&nn->reconfig_lock); + nn->reconfig_sync_present = false; +spin_unlock_bh (&nn->reconfig_lock); +} + +/** +* nfp_net_reconfig() - Reconfigure the firmware +* @nn: NFP Net device to reconfigure +* @update: The value for the update field in the BAR config +* +* Write the update word to the BAR and ping the reconfig queue. The +* poll until the firmware has acknowledged the update by zeroing the +* update word. +* +* Return: Negative errno on error, 0 on success +*/ +int nfp_net_reconfig (struct nfp_net *nn,u32 update) +{ +int ret; + +nfp_net_reconfig_sync_enter (nn);nfp_net_reconfig_start (nn, update); ret =nfp_net_reconfig_wait (nn, jiffies +HZ *NFP_NET_POLL_TIMEOUT );
@@ -3609,6 +3625,7 @@ struct nfp_net *nfp_net_alloc(struct pci*/ void nfp_net_free (struct nfp_net *nn) { +WARN_ON (timer_pending (&nn->reconfig_timer) || nn->reconfig_posted);if (nn->dp.netdev)free_netdev (nn->dp.netdev);else
@@ -3893,4 +3910,5 @@ void nfp_net_clean(struct nfp_net *nn)return ;unregister_netdev (nn->dp.netdev); +nfp_net_reconfig_wait_posted (nn); }
Greg Kroah-Hartman
maintainerPATCH 33/197 sctp: hold transport before accessing its asoc in sctp_transport_get_next
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xin Long <[email protected]>
[ Upstream commit bab1be79a5169ac748d8292b20c86d874022d7ba ]
As Marcelo noticed, in sctp_transport_get_next, it is iterating over transports but then also accessing the association directly, without checking any refcnts before that, which can cause an use-after-free Read.
So fix it by holding transport before accessing the association. With that, sctp_transport_hold calls can be removed in the later places.
--- net/sctp/proc.c | 4 ---- net/sctp/socket.c | 22 +++++++++++++++------- 2 files changed, 15 insertions(+), 11 deletions(-)
net/sctp/proc.c +0 -4
@@ -260,8 +260,6 @@ static int sctp_assocs_seq_show(struct s } transport = (struct sctp_transport *)v; -if (!sctp_transport_hold (transport)) -return 0 ; assoc = transport->asoc; epb = &assoc->base; sk = epb->sk;
@@ -318,8 +316,6 @@ static int sctp_remaddr_seq_show(struct } transport = (struct sctp_transport *)v; -if (!sctp_transport_hold (transport)) -return 0 ; assoc = transport->asoc; list_for_each_entry_rcu(tsp, &assoc->peer.transport_addr_list,
net/sctp/socket.c +15 -7
@@ -4881,9 +4881,14 @@ struct sctp_transport *sctp_transport_gebreak ; } +if (!sctp_transport_hold (t)) +continue ; +if (net_eq (sock_net (t->asoc->base.sk), net) && t->asoc->peer.primary_path == t)break ; + +sctp_transport_put (t); }return t;
@@ -4893,13 +4898,18 @@ struct sctp_transport *sctp_transport_ge struct rhashtable_iter *iter, int pos) { - void *obj = SEQ_START_TOKEN; +struct sctp_transport *t; + +if (!pos) +return SEQ_START_TOKEN ; -while (pos && (obj =sctp_transport_get_next (net, iter)) && - !IS_ERR (obj)) - pos--; +while ((t =sctp_transport_get_next (net, iter)) && !IS_ERR (t)) { +if (!--pos) +break ; +sctp_transport_put (t); + } -return obj; +return t; } int sctp_for_each_endpoint(int (*cb)(struct sctp_endpoint *, void *),
@@ -4958,8 +4968,6 @@ again: tsp =sctp_transport_get_idx (net, &hti, *pos +1 );for (; !IS_ERR_OR_NULL (tsp); tsp =sctp_transport_get_next (net, &hti)) { -if (!sctp_transport_hold (tsp)) -continue ; ret =cb (tsp, p);if (ret)break ;
Greg Kroah-Hartman
maintainerPATCH 34/197 mlxsw: spectrum_switchdev: Do not leak RIFs when removing bridge
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ido Schimmel <[email protected]>
[ Upstream commit 602b74eda81311dbdb5dbab08c30f789f648ebdc ]
When a bridge device is removed, the VLANs are flushed from each configured port. This causes the ports to decrement the reference count on the associated FIDs (filtering identifier). If the reference count of a FID is 1 and it has a RIF (router interface), then this RIF is destroyed.
However, if no port is member in the VLAN for which a RIF exists, then the RIF will continue to exist after the removal of the bridge. To reproduce:
# ip link add name br0 type bridge vlan_filtering 1 # ip link set dev swp1 master br0 # ip link add link br0 name br0.10 type vlan id 10 # ip address add 192.0.2.0/24 dev br0.10 # ip link del dev br0
The RIF associated with br0.10 continues to exist.
Fix this by iterating over all the bridge device uppers when it is destroyed and take care of destroying their RIFs.
--- drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 2 + drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 11 ++++++++ drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c | 20 +++++++++++++++ 3 files changed, 33 insertions(+)
drivers/net/ethernet/mellanox/mlxsw/spectrum.h +2 -0
@@ -433,6 +433,8 @@ mlxsw_sp_netdevice_ipip_ul_event(structvoid mlxsw_sp_port_vlan_router_leave (struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan);void mlxsw_sp_rif_destroy (struct mlxsw_sp_rif *rif); +void mlxsw_sp_rif_destroy_by_dev (struct mlxsw_sp *mlxsw_sp, +struct net_device *dev);/* spectrum_kvdl.c */ int mlxsw_sp_kvdl_init (struct mlxsw_sp *mlxsw_sp);
drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +11 -0
@@ -6228,6 +6228,17 @@ void mlxsw_sp_rif_destroy(struct mlxsw_smlxsw_sp_vr_put (mlxsw_sp, vr); } +void mlxsw_sp_rif_destroy_by_dev (struct mlxsw_sp *mlxsw_sp, +struct net_device *dev) +{ +struct mlxsw_sp_rif *rif; + + rif =mlxsw_sp_rif_find_by_dev (mlxsw_sp, dev); +if (!rif) +return ; +mlxsw_sp_rif_destroy (rif); +} +static void mlxsw_sp_rif_subport_params_init (struct mlxsw_sp_rif_params *params,struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan)
drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +20 -0
@@ -160,6 +160,24 @@ bool mlxsw_sp_bridge_device_is_offloadedreturn !!mlxsw_sp_bridge_device_find (mlxsw_sp->bridge, br_dev); } +static int mlxsw_sp_bridge_device_upper_rif_destroy (struct net_device *dev, +void *data) +{ +struct mlxsw_sp *mlxsw_sp = data; + +mlxsw_sp_rif_destroy_by_dev (mlxsw_sp, dev); +return 0 ; +} + +static void mlxsw_sp_bridge_device_rifs_destroy (struct mlxsw_sp *mlxsw_sp, +struct net_device *dev) +{ +mlxsw_sp_rif_destroy_by_dev (mlxsw_sp, dev); +netdev_walk_all_upper_dev_rcu (dev, + mlxsw_sp_bridge_device_upper_rif_destroy, + mlxsw_sp); +} +static struct mlxsw_sp_bridge_device *mlxsw_sp_bridge_device_create (struct mlxsw_sp_bridge *bridge,struct net_device *br_dev)
@@ -198,6 +216,8 @@ static void mlxsw_sp_bridge_device_destroy(struct mlxsw_sp_bridge *bridge, struct mlxsw_sp_bridge_device *bridge_device) { + mlxsw_sp_bridge_device_rifs_destroy(bridge->mlxsw_sp, + bridge_device->dev);list_del (&bridge_device->list);if (bridge_device->vlan_enabled) bridge->vlan_enabled_exists = false;
Greg Kroah-Hartman
maintainerPATCH 35/197 vhost: correctly check the iova range when waking virtqueue
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jason Wang <[email protected]>
[ Upstream commit 2d66f997f0545c8f7fc5cf0b49af1decb35170e7 ]
We don't wakeup the virtqueue if the first byte of pending iova range is the last byte of the range we just got updated. This will lead a virtqueue to wait for IOTLB updating forever. Fixing by correct the check and wake up the virtqueue in this case.
--- drivers/vhost/vhost.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/vhost/vhost.c +1 -1
@@ -951,7 +951,7 @@ static void vhost_iotlb_notify_vq(structlist_for_each_entry_safe (node, n, &d->pending_list, node) {struct vhost_iotlb_msg *vq_msg = &node->msg.iotlb;if (msg->iova <= vq_msg->iova && - msg->iova + msg->size -1 > vq_msg->iova && + msg->iova + msg->size -1 >= vq_msg->iova && vq_msg->type ==VHOST_IOTLB_MISS ) {vhost_poll_queue (&node->vq->poll);list_del (&node->node);
Greg Kroah-Hartman
maintainerPATCH 36/197 hv_netvsc: ignore devices that are not PCI
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stephen Hemminger <[email protected]>
[ Upstream commit b93c1b5ac8643cc08bb74fa8ae21d6c63dfcb23d ]
Registering another device with same MAC address (such as TAP, VPN or DPDK KNI) will confuse the VF autobinding logic. Restrict the search to only run if the device is known to be a PCI attached VF.
--- drivers/net/hyperv/netvsc_drv.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/net/hyperv/netvsc_drv.c +5 -0
@@ -29,6 +29,7 @@#include <linux/netdevice.h> #include <linux/inetdevice.h> #include <linux/etherdevice.h> +#include <linux/pci.h> #include <linux/skbuff.h> #include <linux/if_vlan.h> #include <linux/in.h>
@@ -1939,12 +1940,16 @@ static int netvsc_register_vf(struct net {struct net_device *ndev;struct net_device_context *net_device_ctx; +struct device *pdev = vf_netdev->dev.parent;struct netvsc_device *netvsc_dev;int ret;if (vf_netdev->addr_len !=ETH_ALEN )return NOTIFY_DONE ; +if (!pdev || !dev_is_pci (pdev) ||dev_is_pf (pdev)) +return NOTIFY_DONE ; +/* * We will use the MAC address to locate the synthetic interface to * associate with the VF interface. If we don't find a matching
Greg Kroah-Hartman
maintainerPATCH 37/197 cifs: check if SMB2 PDU size has been padded and suppress the warning
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ronnie Sahlberg <[email protected]>
[ Upstream commit e6c47dd0da1e3a484e778046fc10da0b20606a86 ]
Some SMB2/3 servers, Win2016 but possibly others too, adds padding not only between PDUs in a compound but also to the final PDU. This padding extends the PDU to a multiple of 8 bytes.
Check if the unexpected length looks like this might be the case and avoid triggering the log messages for :
"SMB2 server sent bad RFC1001 len %d not %d\n"
--- fs/cifs/smb2misc.c | 7 +++++++ 1 file changed, 7 insertions(+)
fs/cifs/smb2misc.c +7 -0
@@ -241,6 +241,13 @@ smb2_check_message(char *buf, unsigned ireturn 0 ;/* +* Some windows servers (win2016) will pad also the final +* PDU in a compound to 8 bytes. +*/ +if (((clc_len +7 ) & ~7 ) == len) +return 0 ; + +/* * MacOS server pads after SMB2.1 write response with 3 bytes * of junk. Other servers match RFC1001 len to actual * SMB2/SMB3 frame length (header + smb2 response specific data)
Greg Kroah-Hartman
maintainerPATCH 38/197 hfsplus: dont return 0 when fill_super() failed
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tetsuo Handa <[email protected]>
[ Upstream commit 7464726cb5998846306ed0a7d6714afb2e37b25d ]
syzbot is reporting NULL pointer dereference at mount_fs() [1]. This is because hfsplus_fill_super() is by error returning 0 when hfsplus_fill_super() detected invalid filesystem image, and mount_bdev() is returning NULL because dget(s->s_root) == NULL if s->s_root == NULL, and mount_fs() is accessing root->d_sb because IS_ERR(root) == false if root == NULL. Fix this by returning -EINVAL when hfsplus_fill_super() detected invalid filesystem image.
[1] https://syzkaller.appspot.com/bug?id=21acb6850cecbc960c927229e597158cf35f33d0
--- fs/hfsplus/super.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
fs/hfsplus/super.c +3 -1
@@ -524,8 +524,10 @@ static int hfsplus_fill_super(struct sup goto out_put_root;if (!hfs_brec_read (&fd, &entry,sizeof (entry))) {hfs_find_exit (&fd); -if (entry.type !=cpu_to_be16 (HFSPLUS_FOLDER )) +if (entry.type !=cpu_to_be16 (HFSPLUS_FOLDER )) { + err = -EINVAL ; goto out_put_root; + } inode =hfsplus_iget (sb,be32_to_cpu (entry.folder.id));if (IS_ERR (inode)) { err =PTR_ERR (inode);
Greg Kroah-Hartman
maintainerPATCH 39/197 hfs: prevent crash on exit from failed search
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Ernesto A. Fern�ndez" <[email protected]>
[ Upstream commit dc2572791d3a41bab94400af2b6bca9d71ccd303 ]
hfs_find_exit() expects fd->bnode to be NULL after a search has failed. hfs_brec_insert() may instead set it to an error-valued pointer. Fix this to prevent a crash.
--- fs/hfs/brec.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
fs/hfs/brec.c +4 -3
@@ -75,9 +75,10 @@ int hfs_brec_insert(struct hfs_find_dataif (!fd->bnode) {if (!tree->root)hfs_btree_inc_height (tree); - fd->bnode =hfs_bnode_find (tree, tree->leaf_head); -if (IS_ERR (fd->bnode)) -return PTR_ERR (fd->bnode); + node =hfs_bnode_find (tree, tree->leaf_head); +if (IS_ERR (node)) +return PTR_ERR (node); + fd->bnode = node; fd->record =-1 ; } new_node =NULL ;
Greg Kroah-Hartman
maintainerPATCH 40/197 sunrpc: Dont use stack buffer with scatterlist
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Laura Abbott <[email protected]>
[ Upstream commit 44090cc876926277329e1608bafc01b9f6da627f ]
Fedora got a bug report from NFS:
kernel BUG at include/linux/scatterlist.h:143!
...
RIP: 0010:sg_init_one+0x7d/0x90 .. make_checksum+0x4e7/0x760 [rpcsec_gss_krb5] gss_get_mic_kerberos+0x26e/0x310 [rpcsec_gss_krb5] gss_marshal+0x126/0x1a0 [auth_rpcgss] ? __local_bh_enable_ip+0x80/0xe0 ? call_transmit_status+0x1d0/0x1d0 [sunrpc] call_transmit+0x137/0x230 [sunrpc] __rpc_execute+0x9b/0x490 [sunrpc] rpc_run_task+0x119/0x150 [sunrpc] nfs4_run_exchange_id+0x1bd/0x250 [nfsv4] _nfs4_proc_exchange_id+0x2d/0x490 [nfsv4] nfs41_discover_server_trunking+0x1c/0xa0 [nfsv4] nfs4_discover_server_trunking+0x80/0x270 [nfsv4] nfs4_init_client+0x16e/0x240 [nfsv4] ? nfs_get_client+0x4c9/0x5d0 [nfs] ? _raw_spin_unlock+0x24/0x30 ? nfs_get_client+0x4c9/0x5d0 [nfs] nfs4_set_client+0xb2/0x100 [nfsv4] nfs4_create_server+0xff/0x290 [nfsv4] nfs4_remote_mount+0x28/0x50 [nfsv4] mount_fs+0x3b/0x16a vfs_kern_mount.part.35+0x54/0x160 nfs_do_root_mount+0x7f/0xc0 [nfsv4] nfs4_try_mount+0x43/0x70 [nfsv4] ? get_nfs_version+0x21/0x80 [nfs] nfs_fs_mount+0x789/0xbf0 [nfs] ? pcpu_alloc+0x6ca/0x7e0 ? nfs_clone_super+0x70/0x70 [nfs] ? nfs_parse_mount_options+0xb40/0xb40 [nfs] mount_fs+0x3b/0x16a vfs_kern_mount.part.35+0x54/0x160 do_mount+0x1fd/0xd50 ksys_mount+0xba/0xd0 __x64_sys_mount+0x21/0x30 do_syscall_64+0x60/0x1f0 entry_SYSCALL_64_after_hwframe+0x49/0xbe
This is BUG_ON(!virt_addr_valid(buf)) triggered by using a stack allocated buffer with a scatterlist. Convert the buffer for rc4salt to be dynamically allocated instead.
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1615258
--- net/sunrpc/auth_gss/gss_krb5_crypto.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
net/sunrpc/auth_gss/gss_krb5_crypto.c +9 -3
@@ -169,7 +169,7 @@ make_checksum_hmac_md5(struct krb5_ctx *struct scatterlist sg[1 ];int err =-1 ;u8 *checksumdata; -u8 rc4salt[4 ]; +u8 *rc4salt;struct crypto_ahash *md5;struct crypto_ahash *hmac_md5;struct ahash_request *req;
@@ -183,14 +183,18 @@ make_checksum_hmac_md5(struct krb5_ctx *return GSS_S_FAILURE ; } + rc4salt =kmalloc_array (4 ,sizeof (*rc4salt),GFP_NOFS ); +if (!rc4salt) +return GSS_S_FAILURE ; +if (arcfour_hmac_md5_usage_to_salt (usage, rc4salt)) {dprintk ("%s: invalid usage value %u\n" , __func__, usage); -return GSS_S_FAILURE ; + goto out_free_rc4salt; } checksumdata =kmalloc (GSS_KRB5_MAX_CKSUM_LEN ,GFP_NOFS );if (!checksumdata) -return GSS_S_FAILURE ; + goto out_free_rc4salt; md5 =crypto_alloc_ahash ("md5" ,0 ,CRYPTO_ALG_ASYNC );if (IS_ERR (md5))
@@ -258,6 +262,8 @@ out_free_md5:crypto_free_ahash (md5); out_free_cksum:kfree (checksumdata); +out_free_rc4salt: +kfree (rc4salt);return err ?GSS_S_FAILURE :0 ; }
Greg Kroah-Hartman
maintainerPATCH 41/197 fork: dont copy inconsistent signal handler state to child
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jann Horn <[email protected]>
[ Upstream commit 06e62a46bbba20aa5286102016a04214bb446141 ]
Before this change, if a multithreaded process forks while one of its threads is changing a signal handler using sigaction(), the memcpy() in copy_sighand() can race with the struct assignment in do_sigaction(). It isn't clear whether this can cause corruption of the userspace signal handler pointer, but it definitely can cause inconsistency between different fields of struct sigaction.
Take the appropriate spinlock to avoid this.
I have tested that this patch prevents inconsistency between sa_sigaction and sa_flags, which is possible before this patch.
--- kernel/fork.c | 2 ++ 1 file changed, 2 insertions(+)
kernel/fork.c +2 -0
@@ -1417,7 +1417,9 @@ static int copy_sighand(unsigned long clreturn -ENOMEM ;atomic_set (&sig->count,1 ); +spin_lock_irq (¤t->sighand->siglock);memcpy (sig->action, current->sighand->action,sizeof (sig->action)); +spin_unlock_irq (¤t->sighand->siglock);return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 42/197 fs/proc/vmcore.c: hide vmcoredd_mmap_dumps() for nommu builds
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit a2036a1ef2ee91acab01a0ae4a534070691a42ec ]
Without CONFIG_MMU, we get a build warning:
fs/proc/vmcore.c:228:12: error: 'vmcoredd_mmap_dumps' defined but not used [-Werror=unused-function]
static int vmcoredd_mmap_dumps(struct vm_area_struct *vma, unsigned long dst,
The function is only referenced from an #ifdef'ed caller, so this uses the same #ifdef around it.
--- fs/proc/vmcore.c | 2 ++ 1 file changed, 2 insertions(+)
fs/proc/vmcore.c +2 -0
@@ -225,6 +225,7 @@ out_unlock:return ret; } +#ifdef CONFIG_MMU static int vmcoredd_mmap_dumps (struct vm_area_struct *vma,unsigned long dst,u64 start,size_t size) {
@@ -259,6 +260,7 @@ out_unlock:mutex_unlock (&vmcoredd_mutex);return ret; } +#endif /* CONFIG_MMU */ #endif /* CONFIG_PROC_VMCORE_DEVICE_DUMP */ /* Read from the ELF header and then the crash dump. On error, negative value is
Greg Kroah-Hartman
maintainerPATCH 43/197 reiserfs: change j_timestamp type to time64_t
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit 8b73ce6a4bae4fe12bcb2c361c0da4183c2e1b6f ]
This uses the deprecated time_t type but is write-only, and could be removed, but as Jeff explains, having a timestamp can be usefule for post-mortem analysis in crash dumps.
In order to remove one of the last instances of time_t, this changes the type to time64_t, same as j_trans_start_time.
--- fs/reiserfs/reiserfs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/reiserfs/reiserfs.h +1 -1
@@ -271,7 +271,7 @@ struct reiserfs_journal_list {struct mutex j_commit_mutex;unsigned int j_trans_id; -time_t j_timestamp; +time64_t j_timestamp;/* write-only but useful for crash dump analysis */ struct reiserfs_list_bitmap *j_list_bitmap;struct buffer_head *j_commit_bh;/* commit buffer head */ struct reiserfs_journal_cnode *j_realblock;
Greg Kroah-Hartman
maintainerPATCH 44/197 iommu/rockchip: Handle errors returned from PM framework
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marc Zyngier <[email protected]>
[ Upstream commit 3fc7c5c0cff3150e471f5fd12f59971c6d2c6513 ]
pm_runtime_get_if_in_use can fail: either PM has been disabled altogether (-EINVAL), or the device hasn't been enabled yet (0). Sadly, the Rockchip IOMMU driver tends to conflate the two things by considering a non-zero return value as successful.
This has the consequence of hiding other bugs, so let's handle this case throughout the driver, with a WARN_ON_ONCE so that we can try and work out what happened.
--- drivers/iommu/rockchip-iommu.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-)
drivers/iommu/rockchip-iommu.c +15 -6
@@ -521,10 +521,11 @@ static irqreturn_t rk_iommu_irq(int irq,u32 int_status;dma_addr_t iova;irqreturn_t ret =IRQ_NONE ; -int i; +int i, err; -if (WARN_ON (!pm_runtime_get_if_in_use (iommu->dev))) -return 0 ; + err =pm_runtime_get_if_in_use (iommu->dev); +if (WARN_ON_ONCE (err <=0 )) +return ret;if (WARN_ON (clk_bulk_enable (iommu->num_clocks, iommu->clocks))) goto out;
@@ -620,11 +621,15 @@ static void rk_iommu_zap_iova(struct rk_spin_lock_irqsave (&rk_domain->iommus_lock, flags);list_for_each (pos, &rk_domain->iommus) {struct rk_iommu *iommu; +int ret; iommu = list_entry(pos, struct rk_iommu, node);/* Only zap TLBs of IOMMUs that are powered on. */ -if (pm_runtime_get_if_in_use (iommu->dev)) { + ret =pm_runtime_get_if_in_use (iommu->dev); +if (WARN_ON_ONCE (ret <0 )) +continue ; +if (ret) {WARN_ON (clk_bulk_enable (iommu->num_clocks, iommu->clocks));rk_iommu_zap_lines (iommu, iova, size);
@@ -891,6 +896,7 @@ static void rk_iommu_detach_device(strucstruct rk_iommu *iommu;struct rk_iommu_domain *rk_domain =to_rk_domain (domain);unsigned long flags; +int ret;/* Allow 'virtual devices' (eg drm) to detach from domain */ iommu =rk_iommu_from_dev (dev);
@@ -909,7 +915,9 @@ static void rk_iommu_detach_device(struclist_del_init (&iommu->node);spin_unlock_irqrestore (&rk_domain->iommus_lock, flags); -if (pm_runtime_get_if_in_use (iommu->dev)) { + ret =pm_runtime_get_if_in_use (iommu->dev); +WARN_ON_ONCE (ret <0 ); +if (ret >0 ) {rk_iommu_disable (iommu);pm_runtime_put (iommu->dev); }
@@ -946,7 +954,8 @@ static int rk_iommu_attach_device(structlist_add_tail (&iommu->node, &rk_domain->iommus);spin_unlock_irqrestore (&rk_domain->iommus_lock, flags); -if (!pm_runtime_get_if_in_use (iommu->dev)) + ret =pm_runtime_get_if_in_use (iommu->dev); +if (!ret ||WARN_ON_ONCE (ret <0 ))return 0 ; ret =rk_iommu_enable (iommu);
Greg Kroah-Hartman
maintainerPATCH 45/197 hfsplus: fix NULL dereference in hfsplus_lookup()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Ernesto A. Fern�ndez" <[email protected]>
[ Upstream commit a7ec7a4193a2eb3b5341243fc0b621c1ac9e4ec4 ]
An HFS+ filesystem can be mounted read-only without having a metadata directory, which is needed to support hardlinks. But if the catalog data is corrupted, a directory lookup may still find dentries claiming to be hardlinks.
hfsplus_lookup() does check that ->hidden_dir is not NULL in such a situation, but mistakenly does so after dereferencing it for the first time. Reorder this check to prevent a crash.
This happens when looking up corrupted catalog data (dentry) on a filesystem with no metadata directory (this could only ever happen on a read-only mount). Wen Xu sent the replication steps in detail to the fsdevel list: https://bugzilla.kernel.org/show_bug.cgi?id=200297
--- fs/hfsplus/dir.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
fs/hfsplus/dir.c +2 -2
@@ -78,13 +78,13 @@ again:cpu_to_be32 (HFSP_HARDLINK_TYPE ) && entry.file.user_info.fdCreator ==cpu_to_be32 (HFSP_HFSPLUS_CREATOR ) && +HFSPLUS_SB (sb)->hidden_dir && (entry.file.create_date ==HFSPLUS_I (HFSPLUS_SB (sb)->hidden_dir)-> create_date || entry.file.create_date ==HFSPLUS_I (d_inode (sb->s_root))-> - create_date) && - HFSPLUS_SB(sb)->hidden_dir) { + create_date)) { struct qstr str;char name[32 ];
Greg Kroah-Hartman
maintainerPATCH 46/197 iommu/rockchip: Move irq request past pm_runtime_enable
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marc Zyngier <[email protected]>
[ Upstream commit 1aa55ca9b14af6cfd987ce4fdaf548f7067a5d07 ]
Enabling the interrupt early, before power has been applied to the device, can result in an interrupt being delivered too early if:
- the IOMMU shares an interrupt with a VOP
- the VOP has a pending interrupt (after a kexec, for example)
In these conditions, we end-up taking the interrupt without the IOMMU being ready to handle the interrupt (not powered on).
Moving the interrupt request past the pm_runtime_enable() call makes sure we can at least access the IOMMU registers. Note that this is only a partial fix, and that the VOP interrupt will still be screaming until the VOP driver kicks in, which advocates for a more synchronized interrupt enabling/disabling approach.
--- drivers/iommu/rockchip-iommu.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-)
drivers/iommu/rockchip-iommu.c +13 -11
@@ -1161,17 +1161,6 @@ static int rk_iommu_probe(struct platforif (iommu->num_mmu ==0 )return PTR_ERR (iommu->bases[0 ]); - i =0 ; -while ((irq =platform_get_irq (pdev, i++)) != -ENXIO ) { -if (irq <0 ) -return irq; - - err =devm_request_irq (iommu->dev, irq, rk_iommu_irq, -IRQF_SHARED ,dev_name (dev), iommu); -if (err) -return err; - } - iommu->reset_disabled =device_property_read_bool (dev,"rockchip,disable-mmu-reset" );
@@ -1228,6 +1217,19 @@ static int rk_iommu_probe(struct platforpm_runtime_enable (dev); + i =0 ; +while ((irq =platform_get_irq (pdev, i++)) != -ENXIO ) { +if (irq <0 ) +return irq; + + err =devm_request_irq (iommu->dev, irq, rk_iommu_irq, +IRQF_SHARED ,dev_name (dev), iommu); +if (err) { +pm_runtime_disable (dev); + goto err_remove_sysfs; + } + } +return 0 ; err_remove_sysfs:iommu_device_sysfs_remove (&iommu->iommu);
Greg Kroah-Hartman
maintainerPATCH 47/197 fs/proc/kcore.c: use __pa_symbol() for KCORE_TEXT list entries
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: James Morse <[email protected]>
[ Upstream commit df865e8337c397471b95f51017fea559bc8abb4a ]
elf_kcore_store_hdr() uses __pa() to find the physical address of KCORE_RAM or KCORE_TEXT entries exported as program headers.
This trips CONFIG_DEBUG_VIRTUAL's checks, as the KCORE_TEXT entries are not in the linear map.
Handle these two cases separately, using __pa_symbol() for the KCORE_TEXT entries.
--- fs/proc/kcore.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
fs/proc/kcore.c +3 -1
@@ -384,8 +384,10 @@ static void elf_kcore_store_hdr(char *bu phdr->p_flags =PF_R |PF_W |PF_X ; phdr->p_offset =kc_vaddr_to_offset (m->addr) + dataoff; phdr->p_vaddr = (size_t )m->addr; -if (m->type ==KCORE_RAM || m->type ==KCORE_TEXT ) +if (m->type ==KCORE_RAM ) phdr->p_paddr =__pa (m->addr); +else if (m->type ==KCORE_TEXT ) + phdr->p_paddr =__pa_symbol (m->addr);else phdr->p_paddr = (elf_addr_t)-1 ; phdr->p_filesz = phdr->p_memsz = m->size;
Greg Kroah-Hartman
maintainerPATCH 48/197 fat: validate ->i_start before using
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: OGAWA Hirofumi <[email protected]>
[ Upstream commit 0afa9626667c3659ef8bd82d42a11e39fedf235c ]
On corrupted FATfs may have invalid ->i_start. To handle it, this checks ->i_start before using, and return proper error code.
--- fs/fat/cache.c | 19 ++++++++++++------- fs/fat/fat.h | 5 +++++ fs/fat/fatent.c | 6 +++--- 3 files changed, 20 insertions(+), 10 deletions(-)
fs/fat/cache.c +12 -7
@@ -225,7 +225,8 @@ static inline void cache_init(struct fatint fat_get_cluster (struct inode *inode,int cluster,int *fclus,int *dclus) {struct super_block *sb = inode->i_sb; -const int limit = sb->s_maxbytes >>MSDOS_SB (sb)->cluster_bits; +struct msdos_sb_info *sbi =MSDOS_SB (sb); +const int limit = sb->s_maxbytes >> sbi->cluster_bits;struct fat_entry fatent;struct fat_cache_id cid;int nr;
@@ -234,6 +235,12 @@ int fat_get_cluster(struct inode *inode, *fclus =0 ; *dclus =MSDOS_I (inode)->i_start; +if (!fat_valid_entry (sbi, *dclus)) { +fat_fs_error_ratelimit (sb, +"%s: invalid start cluster (i_pos %lld, start %08x)" , + __func__,MSDOS_I (inode)->i_pos, *dclus); +return -EIO ; + }if (cluster ==0 )return 0 ;
@@ -250,9 +257,8 @@ int fat_get_cluster(struct inode *inode,/* prevent the infinite loop of cluster chain */ if (*fclus > limit) {fat_fs_error_ratelimit (sb, -"%s: detected the cluster chain loop" -" (i_pos %lld)" , __func__, -MSDOS_I (inode)->i_pos); +"%s: detected the cluster chain loop (i_pos %lld)" , + __func__,MSDOS_I (inode)->i_pos); nr = -EIO ; goto out; }
@@ -262,9 +268,8 @@ int fat_get_cluster(struct inode *inode, goto out; else if (nr == FAT_ENT_FREE) {fat_fs_error_ratelimit (sb, -"%s: invalid cluster chain (i_pos %lld)" , - __func__, -MSDOS_I (inode)->i_pos); +"%s: invalid cluster chain (i_pos %lld)" , + __func__,MSDOS_I (inode)->i_pos); nr = -EIO ; goto out; } else if (nr == FAT_ENT_EOF) {
fs/fat/fat.h +5 -0
@@ -348,6 +348,11 @@ static inline void fatent_brelse(struct fatent->fat_inode =NULL ; } +static inline bool fat_valid_entry (struct msdos_sb_info *sbi,int entry) +{ +return FAT_START_ENT <= entry && entry < sbi->max_cluster; +} +extern void fat_ent_access_init (struct super_block *sb);extern int fat_ent_read (struct inode *inode,struct fat_entry *fatent,int entry);
fs/fat/fatent.c +3 -3
@@ -23,7 +23,7 @@ static void fat12_ent_blocknr(struct sup {struct msdos_sb_info *sbi =MSDOS_SB (sb);int bytes = entry + (entry >>1 ); -WARN_ON (entry <FAT_START_ENT || sbi->max_cluster <= entry); +WARN_ON (!fat_valid_entry (sbi, entry)); *offset = bytes & (sb->s_blocksize -1 ); *blocknr = sbi->fat_start + (bytes >> sb->s_blocksize_bits); }
@@ -33,7 +33,7 @@ static void fat_ent_blocknr(struct super {struct msdos_sb_info *sbi =MSDOS_SB (sb);int bytes = (entry << sbi->fatent_shift); -WARN_ON (entry <FAT_START_ENT || sbi->max_cluster <= entry); +WARN_ON (!fat_valid_entry (sbi, entry)); *offset = bytes & (sb->s_blocksize -1 ); *blocknr = sbi->fat_start + (bytes >> sb->s_blocksize_bits); }
@@ -353,7 +353,7 @@ int fat_ent_read(struct inode *inode, stint err, offset;sector_t blocknr; -if (entry <FAT_START_ENT || sbi->max_cluster <= entry) { +if (!fat_valid_entry (sbi, entry)) {fatent_brelse (fatent);fat_fs_error (sb,"invalid access to FAT (entry 0x%08x)" , entry);return -EIO ;
Greg Kroah-Hartman
maintainerPATCH 49/197 workqueue: skip lockdep wq dependency in cancel_work_sync()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johannes Berg <[email protected]>
[ Upstream commit d6e89786bed977f37f55ffca11e563f6d2b1e3b5 ]
In cancel_work_sync(), we can only have one of two cases, even with an ordered workqueue:
* the work isn't running, just cancelled before it started
* the work is running, but then nothing else can be on the workqueue before it
Thus, we need to skip the lockdep workqueue dependency handling, otherwise we get false positive reports from lockdep saying that we have a potential deadlock when the workqueue also has other work items with locking, e.g.
work1_function() { mutex_lock(&mutex); ... }
work2_function() { /* nothing */ }
other_function() {
queue_work(ordered_wq, &work1);
queue_work(ordered_wq, &work2);
mutex_lock(&mutex);
cancel_work_sync(&work2);
}As described above, this isn't a problem, but lockdep will currently flag it as if cancel_work_sync() was flush_work(), which *is* a problem.
--- kernel/workqueue.c | 37 ++++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-)
kernel/workqueue.c +22 -15
@@ -2843,7 +2843,8 @@ reflush: }EXPORT_SYMBOL_GPL (drain_workqueue); -static bool start_flush_work (struct work_struct *work,struct wq_barrier *barr) +static bool start_flush_work (struct work_struct *work,struct wq_barrier *barr, +bool from_cancel) {struct worker *worker =NULL ;struct worker_pool *pool;
@@ -2885,7 +2886,8 @@ static bool start_flush_work(struct work* workqueues the deadlock happens when the rescuer stalls, blocking * forward progress. */ -if (pwq->wq->saved_max_active ==1 || pwq->wq->rescuer) { +if (!from_cancel && + (pwq->wq->saved_max_active ==1 || pwq->wq->rescuer)) {lock_map_acquire (&pwq->wq->lockdep_map);lock_map_release (&pwq->wq->lockdep_map); }
@@ -2896,6 +2898,22 @@ already_gone:return false; } +static bool __flush_work (struct work_struct *work,bool from_cancel) +{ +struct wq_barrier barr; + +if (WARN_ON (!wq_online)) +return false; + +if (start_flush_work (work, &barr, from_cancel)) { +wait_for_completion (&barr.done); +destroy_work_on_stack (&barr.work); +return true; + }else { +return false; + } +} +/** * flush_work - wait for a work to finish executing the last queueing instance * @work: the work to flush
@@ -2909,18 +2927,7 @@ already_gone:*/ bool flush_work (struct work_struct *work) { -struct wq_barrier barr; - -if (WARN_ON (!wq_online)) -return false; - -if (start_flush_work (work, &barr)) { -wait_for_completion (&barr.done); -destroy_work_on_stack (&barr.work); -return true; - }else { -return false; - } +return __flush_work (work, false); }EXPORT_SYMBOL_GPL (flush_work);
@@ -2986,7 +2993,7 @@ static bool __cancel_work_timer(struct w* isn't executing. */ if (wq_online) -flush_work (work); +__flush_work (work, true);clear_work_data (work);
Greg Kroah-Hartman
maintainerPATCH 50/197 workqueue: re-add lockdep dependencies for flushing
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johannes Berg <[email protected]>
[ Upstream commit 87915adc3f0acdf03c776df42e308e5a155c19af ]
In flush_work(), we need to create a lockdep dependency so that the following scenario is appropriately tagged as a problem:
work_function()
{
mutex_lock(&mutex);
...
}
other_function()
{
mutex_lock(&mutex);
flush_work(&work); // or cancel_work_sync(&work);
}This is a problem since the work might be running and be blocked on trying to acquire the mutex.
Similarly, in flush_workqueue().
These were removed after cross-release partially caught these problems, but now cross-release was reverted anyway. IMHO the removal was erroneous anyway though, since lockdep should be able to catch potential problems, not just actual ones, and cross-release would only have caught the problem when actually invoking wait_for_completion().
--- kernel/workqueue.c | 8 ++++++++ 1 file changed, 8 insertions(+)
kernel/workqueue.c +8 -0
@@ -2652,6 +2652,9 @@ void flush_workqueue(struct workqueue_stif (WARN_ON (!wq_online))return ; +lock_map_acquire (&wq->lockdep_map); +lock_map_release (&wq->lockdep_map); +mutex_lock (&wq->mutex);/*
@@ -2905,6 +2908,11 @@ static bool __flush_work(struct work_strif (WARN_ON (!wq_online))return false; +if (!from_cancel) { +lock_map_acquire (&work->lockdep_map); +lock_map_release (&work->lockdep_map); + } +if (start_flush_work (work, &barr, from_cancel)) {wait_for_completion (&barr.done);destroy_work_on_stack (&barr.work);
Greg Kroah-Hartman
maintainerPATCH 51/197 scripts: modpost: check memory allocation results
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Randy Dunlap <[email protected]>
[ Upstream commit 1f3aa9002dc6a0d59a4b599b4fc8f01cf43ef014 ]
Fix missing error check for memory allocation functions in scripts/mod/modpost.c.
Fixes kernel bugzilla #200319:
https://bugzilla.kernel.org/show_bug.cgi?id=200319
--- scripts/mod/modpost.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
scripts/mod/modpost.c +4 -4
@@ -672,7 +672,7 @@ static void handle_modversions(struct moif (ELF_ST_TYPE (sym->st_info) ==STT_SPARC_REGISTER )break ;if (symname[0 ] =='.' ) { -char *munged =strdup (symname); +char *munged =NOFAIL (strdup (symname)); munged[0 ] ='_' ; munged[1 ] =toupper (munged[1 ]); symname = munged;
@@ -1318,7 +1318,7 @@ static Elf_Sym *find_elf_symbol2(structstatic char *sec2annotation (const char *s) {if (match (s, init_exit_sections)) { -char *p =malloc (20 ); +char *p =NOFAIL (malloc (20 ));char *r = p; *p++ ='_' ;
@@ -1338,7 +1338,7 @@ static char *sec2annotation(const char *strcat (p," " );return r; }else { -return strdup ("" ); +return NOFAIL (strdup ("" )); } }
@@ -2036,7 +2036,7 @@ void buf_write(struct buffer *buf, const {if (buf->size - buf->pos < len) { buf->size += len +SZ ; - buf->p =realloc (buf->p, buf->size); + buf->p =NOFAIL (realloc (buf->p, buf->size)); }strncpy (buf->p + buf->pos, s, len); buf->pos += len;
Greg Kroah-Hartman
maintainerPATCH 52/197 apparmor: fix an error code in __aa_create_ns()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dan Carpenter <[email protected]>
[ Upstream commit 0a6b29230ec336189bab32498df3f06c8a6944d8 ]
We should return error pointers in this function. Returning NULL results in a NULL dereference in the caller.
--- security/apparmor/policy_ns.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
security/apparmor/policy_ns.c +1 -1
@@ -255,7 +255,7 @@ static struct aa_ns *__aa_create_ns(stru ns =alloc_ns (parent->base.hname, name);if (!ns) -return NULL ; +return ERR_PTR (-ENOMEM ); ns->level = parent->level +1 ;mutex_lock_nested (&ns->lock, ns->level); error =__aafs_ns_mkdir (ns,ns_subns_dir (parent), name, dir);
Greg Kroah-Hartman
maintainerPATCH 53/197 virtio: pci-legacy: Validate queue pfn
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Suzuki K Poulose <[email protected]>
[ Upstream commit 69599206ea9a3f8f2e94d46580579cbf9d08ad6c ]
Legacy PCI over virtio uses a 32bit PFN for the queue. If the queue pfn is too large to fit in 32bits, which we could hit on arm64 systems with 52bit physical addresses (even with 64K page size), we simply miss out a proper link to the other side of the queue.
Add a check to validate the PFN, rather than silently breaking the devices.
--- drivers/virtio/virtio_pci_legacy.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/virtio/virtio_pci_legacy.c +12 -2
@@ -122,6 +122,7 @@ static struct virtqueue *setup_vq(structstruct virtqueue *vq;u16 num;int err; +u64 q_pfn;/* Select the queue we're interested in */ iowrite16 (index, vp_dev->ioaddr +VIRTIO_PCI_QUEUE_SEL );
@@ -141,9 +142,17 @@ static struct virtqueue *setup_vq(structif (!vq)return ERR_PTR (-ENOMEM ); + q_pfn =virtqueue_get_desc_addr (vq) >>VIRTIO_PCI_QUEUE_ADDR_SHIFT ; +if (q_pfn >>32 ) { +dev_err (&vp_dev->pci_dev->dev, +"platform bug: legacy virtio-mmio must not be used with RAM above 0x%llxGB\n" , +0x1ULL << (32 +PAGE_SHIFT -30 )); + err = -E2BIG ; + goto out_del_vq; + } +/* activate the queue */ -iowrite32 (virtqueue_get_desc_addr (vq) >>VIRTIO_PCI_QUEUE_ADDR_SHIFT , - vp_dev->ioaddr +VIRTIO_PCI_QUEUE_PFN ); +iowrite32 (q_pfn, vp_dev->ioaddr +VIRTIO_PCI_QUEUE_PFN ); vq->priv = (void __force *)vp_dev->ioaddr +VIRTIO_PCI_QUEUE_NOTIFY ;
@@ -160,6 +169,7 @@ static struct virtqueue *setup_vq(struct out_deactivate:iowrite32 (0 , vp_dev->ioaddr +VIRTIO_PCI_QUEUE_PFN ); +out_del_vq:vring_del_virtqueue (vq);return ERR_PTR (err); }
Greg Kroah-Hartman
maintainerPATCH 54/197 x86/mce: Add notifier_block forward declaration
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit 704ae091b061082b37a9968621af4c290c641d50 ]
Without linux/irq.h, there is no declaration of notifier_block, leading to a build warning:
In file included from arch/x86/kernel/cpu/mcheck/threshold.c:10:arch/x86/include/asm/mce.h:151:46: error: 'struct notifier_block' declared inside parameter list will not be visible outside of this definition or declaration [-Werror]
It's sufficient to declare the struct tag here, which avoids pulling in more header files.
--- arch/x86/include/asm/mce.h | 1 + 1 file changed, 1 insertion(+)
arch/x86/include/asm/mce.h +1 -0
@@ -148,6 +148,7 @@ enum mce_notifier_prios { MCE_PRIO_LOWEST = 0, }; +struct notifier_block ;extern void mce_register_decode_chain (struct notifier_block *nb);extern void mce_unregister_decode_chain (struct notifier_block *nb);
Greg Kroah-Hartman
maintainerPATCH 55/197 i2c: core: ACPI: Make acpi_gsb_i2c_read_bytes() check i2c_transfer return value
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hans de Goede <[email protected]>
[ Upstream commit 0a30446c0dca3483c384b54a431cc951e15f7e79 ]
Currently acpi_gsb_i2c_read_bytes() directly returns i2c_transfer's return value. i2c_transfer returns a value < 0 on error and 2 (for 2 successfully executed transfers) on success. But the ACPI code expects 0 on success, so currently acpi_gsb_i2c_read_bytes()'s caller does:
if (status > 0)
status = 0;This commit makes acpi_gsb_i2c_read_bytes() return a value which can be directly consumed by the ACPI code, mirroring acpi_gsb_i2c_write_bytes(), this commit also makes acpi_gsb_i2c_read_bytes() explitcly check that i2c_transfer returns 2, rather then accepting any value > 0.
--- drivers/i2c/i2c-core-acpi.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/i2c/i2c-core-acpi.c +5 -3
@@ -453,8 +453,12 @@ static int acpi_gsb_i2c_read_bytes(strucelse dev_err (&client->adapter->dev,"i2c read %d bytes from client@%#x starting at reg %#x failed, error: %d\n" , data_len, client->addr, cmd, ret); - }else { +/* 2 transfers must have completed successfully */ + }else if (ret ==2 ) {memcpy (data, buffer, data_len); + ret =0 ; + }else { + ret = -EIO ; }kfree (buffer);
@@ -595,8 +599,6 @@ i2c_acpi_space_handler(u32 function, acpif (action ==ACPI_READ ) { status =acpi_gsb_i2c_read_bytes (client, command, gsb->data, info->access_length); -if (status >0 ) - status =0 ; }else { status =acpi_gsb_i2c_write_bytes (client, command, gsb->data, info->access_length);
Greg Kroah-Hartman
maintainerPATCH 56/197 IB/hfi1: Invalid NUMA node information can cause a divide by zero
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Michael J. Ruhl" <[email protected]>
[ Upstream commit c513de490f808d8480346f9a58e6a4a5f3de12e7 ]
If the system BIOS does not supply NUMA node information to the PCI devices, the NUMA node is selected by choosing the current node.
This can lead to the following crash:
divide error: 0000 SMP
CPU: 0 PID: 4 Comm: kworker/0:0 Tainted: G IOE ------------ 3.10.0-693.21.1.el7.x86_64 #1 Hardware name: Intel Corporation S2600KP/S2600KP, BIOS SE5C610.86B.01.01.0005.101720141054 10/17/2014 Workqueue: events work_for_cpu_fn task: ffff880174480fd0 ti: ffff880174488000 task.ti: ffff880174488000 RIP: 0010: [<ffffffffc020ac69>] hfi1_dev_affinity_init+0x129/0x6a0 [hfi1] RSP: 0018:ffff88017448bbf8 EFLAGS: 00010246 RAX: 0000000000000011 RBX: ffff88107ffba6c0 RCX: ffff88085c22e130 RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff880824ad0000 RBP: ffff88017448bc48 R08: 0000000000000011 R09: 0000000000000002 R10: ffff8808582b6ca0 R11: 0000000000003151 R12: ffff8808582b6ca0 R13: ffff8808582b6518 R14: ffff8808582b6010 R15: 0000000000000012 FS: 0000000000000000(0000) GS:ffff88085ec00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007efc707404f0 CR3: 0000000001a02000 CR4: 00000000001607f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Call Trace: hfi1_init_dd+0x14b3/0x27a0 [hfi1] ? pcie_capability_write_word+0x46/0x70 ? hfi1_pcie_init+0xc0/0x200 [hfi1] do_init_one+0x153/0x4c0 [hfi1] ? sched_clock_cpu+0x85/0xc0 init_one+0x1b5/0x260 [hfi1] local_pci_probe+0x4a/0xb0 work_for_cpu_fn+0x1a/0x30 process_one_work+0x17f/0x440 worker_thread+0x278/0x3c0 ? manage_workers.isra.24+0x2a0/0x2a0 kthread+0xd1/0xe0 ? insert_kthread_work+0x40/0x40 ret_from_fork+0x77/0xb0 ? insert_kthread_work+0x40/0x40
If the BIOS is not supplying NUMA information:
- set the default table count to 1 for all possible nodes
- select node 0 (instead of current NUMA) node to get consistent performance
- generate an error indicating that the BIOS should be upgraded
--- drivers/infiniband/hw/hfi1/affinity.c | 24 +++++++++++++++++++++--- 1 file changed, 21 insertions(+), 3 deletions(-)
drivers/infiniband/hw/hfi1/affinity.c +21 -3
@@ -198,7 +198,7 @@ int node_affinity_init(void)while ((dev =pci_get_device (ids->vendor, ids->device, dev))) { node =pcibus_to_node (dev->bus);if (node <0 ) - node =numa_node_id (); + goto out; hfi1_per_node_cntr[node]++; }
@@ -206,6 +206,18 @@ int node_affinity_init(void) }return 0 ; + +out: +/* +* Invalid PCI NUMA node information found, note it, and populate +* our database 1:1. +*/ +pr_err ("HFI: Invalid PCI NUMA node. Performance may be affected\n" ); +pr_err ("HFI: System BIOS may need to be upgraded\n" ); +for (node =0 ; node < node_affinity.num_possible_nodes; node++) + hfi1_per_node_cntr[node] =1 ; + +return 0 ; }static void node_affinity_destroy (struct hfi1_affinity_node *entry)
@@ -622,8 +634,14 @@ int hfi1_dev_affinity_init(struct hfi1_dint curr_cpu, possible, i, ret;bool new_entry = false; -if (node <0 ) - node =numa_node_id (); +/* +* If the BIOS does not have the NUMA node information set, select +* NUMA 0 so we get consistent performance. +*/ +if (node <0 ) { +dd_dev_err (dd,"Invalid PCI NUMA node. Performance may be affected\n" ); + node =0 ; + } dd->node = node; local_mask =cpumask_of_node (dd->node);
Greg Kroah-Hartman
maintainerPATCH 57/197 pwm: meson: Fix mux clock names
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jerome Brunet <[email protected]>
[ Upstream commit b96e9eb62841c519ba1db32d036628be3cdef91f ]
Current clock name looks like this:
/soc/bus@ffd00000/pwm@1b000#mux0
This is bad because CCF uses the clock to create a directory in clk debugfs. With such name, the directory creation (silently) fails and the debugfs entry end up being created at the debugfs root.
With this change, the clock name will now be:
ffd1b000.pwm#mux0
This matches the clock naming scheme used in the ethernet and mmc driver. It also fixes the problem with debugfs.
--- drivers/pwm/pwm-meson.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/pwm/pwm-meson.c +1 -2
@@ -458,7 +458,6 @@ static int meson_pwm_init_channels(struc struct meson_pwm_channel *channels) {struct device *dev = meson->chip.dev; -struct device_node *np = dev->of_node;struct clk_init_data init;unsigned int i;char name[255 ];
@@ -467,7 +466,7 @@ static int meson_pwm_init_channels(strucfor (i =0 ; i < meson->chip.npwm; i++) {struct meson_pwm_channel *channel = &channels[i]; -snprintf (name,sizeof (name),"%pOF#mux%u" , np, i); +snprintf (name,sizeof (name),"%s#mux%u" ,dev_name (dev), i); init.name = name; init.ops = &clk_mux_ops;
Greg Kroah-Hartman
maintainerPATCH 58/197 powerpc/topology: Get topology for shared processors at boot
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Srikar Dronamraju <[email protected]>
[ Upstream commit 2ea62630681027c455117aa471ea3ab8bb099ead ]
On a shared LPAR, Phyp will not update the CPU associativity at boot time. Just after the boot system does recognize itself as a shared LPAR and trigger a request for correct CPU associativity. But by then the scheduler would have already created/destroyed its sched domains.
This causes
- Broken load balance across Nodes causing islands of cores.
- Performance degradation esp if the system is lightly loaded
- dmesg to wrongly report all CPUs to be in Node 0.
- Messages in dmesg saying borken topology.
- With commit 051f3ca02e46 ("sched/topology: Introduce NUMA identity node sched domain"), can cause rcu stalls at boot up.
The sched_domains_numa_masks table which is used to generate cpumasks is only created at boot time just before creating sched domains and never updated. Hence, its better to get the topology correct before the sched domains are created.
For example on 64 core Power 8 shared LPAR, dmesg reports
Brought up 512 CPUs
Node 0 CPUs: 0-511
Node 1 CPUs:
Node 2 CPUs:
Node 3 CPUs:
Node 4 CPUs:
Node 5 CPUs:
Node 6 CPUs:
Node 7 CPUs:
Node 8 CPUs:
Node 9 CPUs:
Node 10 CPUs:
Node 11 CPUs:
...
BUG: arch topology borken
the DIE domain not a subset of the NUMA domain
BUG: arch topology borken
the DIE domain not a subset of the NUMA domainnumactl/lscpu output will still be correct with cores spreading across all nodes:
Socket(s): 64 NUMA node(s): 12 Model: 2.0 (pvr 004d 0200) Model name: POWER8 (architected), altivec supported Hypervisor vendor: pHyp Virtualization type: para L1d cache: 64K L1i cache: 32K NUMA node0 CPU(s): 0-7,32-39,64-71,96-103,176-183,272-279,368-375,464-471 NUMA node1 CPU(s): 8-15,40-47,72-79,104-111,184-191,280-287,376-383,472-479 NUMA node2 CPU(s): 16-23,48-55,80-87,112-119,192-199,288-295,384-391,480-487 NUMA node3 CPU(s): 24-31,56-63,88-95,120-127,200-207,296-303,392-399,488-495 NUMA node4 CPU(s): 208-215,304-311,400-407,496-503 NUMA node5 CPU(s): 168-175,264-271,360-367,456-463 NUMA node6 CPU(s): 128-135,224-231,320-327,416-423 NUMA node7 CPU(s): 136-143,232-239,328-335,424-431 NUMA node8 CPU(s): 216-223,312-319,408-415,504-511 NUMA node9 CPU(s): 144-151,240-247,336-343,432-439 NUMA node10 CPU(s): 152-159,248-255,344-351,440-447 NUMA node11 CPU(s): 160-167,256-263,352-359,448-455
Currently on this LPAR, the scheduler detects 2 levels of Numa and created numa sched domains for all CPUs, but it finds a single DIE domain consisting of all CPUs. Hence it deletes all numa sched domains.
To address this, detect the shared processor and update topology soon after CPUs are setup so that correct topology is updated just before scheduler creates sched domain.
With the fix, dmesg reports:
numa: Node 0 CPUs: 0-7 32-39 64-71 96-103 176-183 272-279 368-375 464-471
numa: Node 1 CPUs: 8-15 40-47 72-79 104-111 184-191 280-287 376-383 472-479
numa: Node 2 CPUs: 16-23 48-55 80-87 112-119 192-199 288-295 384-391 480-487
numa: Node 3 CPUs: 24-31 56-63 88-95 120-127 200-207 296-303 392-399 488-495
numa: Node 4 CPUs: 208-215 304-311 400-407 496-503
numa: Node 5 CPUs: 168-175 264-271 360-367 456-463
numa: Node 6 CPUs: 128-135 224-231 320-327 416-423
numa: Node 7 CPUs: 136-143 232-239 328-335 424-431
numa: Node 8 CPUs: 216-223 312-319 408-415 504-511
numa: Node 9 CPUs: 144-151 240-247 336-343 432-439
numa: Node 10 CPUs: 152-159 248-255 344-351 440-447
numa: Node 11 CPUs: 160-167 256-263 352-359 448-455
and lscpu also reports:
Socket(s): 64 NUMA node(s): 12 Model: 2.0 (pvr 004d 0200) Model name: POWER8 (architected), altivec supported Hypervisor vendor: pHyp Virtualization type: para L1d cache: 64K L1i cache: 32K NUMA node0 CPU(s): 0-7,32-39,64-71,96-103,176-183,272-279,368-375,464-471 NUMA node1 CPU(s): 8-15,40-47,72-79,104-111,184-191,280-287,376-383,472-479 NUMA node2 CPU(s): 16-23,48-55,80-87,112-119,192-199,288-295,384-391,480-487 NUMA node3 CPU(s): 24-31,56-63,88-95,120-127,200-207,296-303,392-399,488-495 NUMA node4 CPU(s): 208-215,304-311,400-407,496-503 NUMA node5 CPU(s): 168-175,264-271,360-367,456-463 NUMA node6 CPU(s): 128-135,224-231,320-327,416-423 NUMA node7 CPU(s): 136-143,232-239,328-335,424-431 NUMA node8 CPU(s): 216-223,312-319,408-415,504-511 NUMA node9 CPU(s): 144-151,240-247,336-343,432-439 NUMA node10 CPU(s): 152-159,248-255,344-351,440-447 NUMA node11 CPU(s): 160-167,256-263,352-359,448-455
--- arch/powerpc/include/asm/topology.h | 5 +++++ arch/powerpc/kernel/smp.c | 5 +++++ arch/powerpc/mm/numa.c | 20 ++++++++++---------- 3 files changed, 20 insertions(+), 10 deletions(-)
arch/powerpc/include/asm/topology.h +5 -0
@@ -92,6 +92,7 @@ extern int stop_topology_update(void);extern int prrn_is_enabled (void );extern int find_and_online_cpu_nid (int cpu);extern int timed_topology_update (int nsecs); +extern void __init shared_proc_topology_init (void );#else static inline int start_topology_update (void ) {
@@ -113,6 +114,10 @@ static inline int timed_topology_update( {return 0 ; } + +#ifdef CONFIG_SMP +static inline void shared_proc_topology_init (void ) {} +#endif #endif /* CONFIG_NUMA && CONFIG_PPC_SPLPAR */ #include <asm-generic/topology.h>
arch/powerpc/kernel/smp.c +5 -0
@@ -1156,6 +1156,11 @@ void __init smp_cpus_done(unsigned int mif (smp_ops && smp_ops->bringup_done) smp_ops->bringup_done (); +/* +* On a shared LPAR, associativity needs to be requested. +* Hence, get numa topology before dumping cpu topology +*/ +shared_proc_topology_init ();dump_numa_cpu_topology ();/*
arch/powerpc/mm/numa.c +10 -10
@@ -1078,7 +1078,6 @@ static int prrn_enabled;static void reset_topology_timer (void );static int topology_timer_secs =1 ;static int topology_inited; -static int topology_update_needed;/* * Change polling interval for associativity changes.
@@ -1306,11 +1305,8 @@ int numa_update_cpu_topology(bool cpus_lstruct device *dev;int weight, new_nid, i =0 ; -if (!prrn_enabled && !vphn_enabled) { -if (!topology_inited) - topology_update_needed =1 ; +if (!prrn_enabled && !vphn_enabled && topology_inited)return 0 ; - } weight =cpumask_weight (&cpu_associativity_changes_mask);if (!weight)
@@ -1423,7 +1419,6 @@ int numa_update_cpu_topology(bool cpus_l out:kfree (updates); - topology_update_needed =0 ;return changed; }
@@ -1551,6 +1546,15 @@ int prrn_is_enabled(void)return prrn_enabled; } +void __init shared_proc_topology_init (void ) +{ +if (lppaca_shared_proc (get_lppaca ())) { +bitmap_fill (cpumask_bits (&cpu_associativity_changes_mask), + nr_cpumask_bits); +numa_update_cpu_topology (false); + } +} +static int topology_read (struct seq_file *file,void *v) {if (vphn_enabled || prrn_enabled)
@@ -1608,10 +1612,6 @@ static int topology_update_init(void)return -ENOMEM ; topology_inited =1 ; -if (topology_update_needed) -bitmap_fill (cpumask_bits (&cpu_associativity_changes_mask), - nr_cpumask_bits); -return 0 ; }device_initcall (topology_update_init);
Greg Kroah-Hartman
maintainerPATCH 59/197 mm/fadvise.c: fix signed overflow UBSAN complaint
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andrey Ryabinin <[email protected]>
[ Upstream commit a718e28f538441a3b6612da9ff226973376cdf0f ]
Signed integer overflow is undefined according to the C standard. The overflow in ksys_fadvise64_64() is deliberate, but since it is signed overflow, UBSAN complains:
UBSAN: Undefined behaviour in mm/fadvise.c:76:10 signed integer overflow: 4 + 9223372036854775805 cannot be represented in type 'long long int'
Use unsigned types to do math. Unsigned overflow is defined so UBSAN will not complain about it. This patch doesn't change generated code.
--- mm/fadvise.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
mm/fadvise.c +6 -2
@@ -72,8 +72,12 @@ int ksys_fadvise64_64(int fd, loff_t off goto out; } -/* Careful about overflows. Len == 0 means "as much as possible" */ - endbyte = offset + len; +/* +* Careful about overflows. Len == 0 means "as much as possible". Use +* unsigned math because signed overflows are undefined and UBSan +* complains. +*/ + endbyte = (u64 )offset + (u64 )len;if (!len || endbyte < len) endbyte =-1 ;else
Greg Kroah-Hartman
maintainerPATCH 60/197 mm: make DEFERRED_STRUCT_PAGE_INIT explicitly depend on SPARSEMEM
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mike Rapoport <[email protected]>
[ Upstream commit d39f8fb4b7776dcb09ec3bf7a321547083078ee3 ]
The deferred memory initialization relies on section definitions, e.g PAGES_PER_SECTION, that are only available when CONFIG_SPARSEMEM=y on most architectures.
Initially DEFERRED_STRUCT_PAGE_INIT depended on explicit ARCH_SUPPORTS_DEFERRED_STRUCT_PAGE_INIT configuration option, but since the commit 2e3ca40f03bb13709df4 ("mm: relax deferred struct page requirements") this requirement was relaxed and now it is possible to enable DEFERRED_STRUCT_PAGE_INIT on architectures that support DISCONTINGMEM and NO_BOOTMEM which causes build failures.
For instance, setting SMP=y and DEFERRED_STRUCT_PAGE_INIT=y on arc causes the following build failure:
CC mm/page_alloc.o
mm/page_alloc.c: In function 'update_defer_init':
mm/page_alloc.c:321:14: error: 'PAGES_PER_SECTION' undeclared (first use in this function); did you mean 'USEC_PER_SEC'?
(pfn & (PAGES_PER_SECTION - 1)) == 0) {
^~~~~~~~~~~~~~~~~
USEC_PER_SEC
mm/page_alloc.c:321:14: note: each undeclared identifier is reported only once for each function it appears in
In file included from include/linux/cache.h:5:0,
from include/linux/printk.h:9,
from include/linux/kernel.h:14,
from include/asm-generic/bug.h:18,
from arch/arc/include/asm/bug.h:32,
from include/linux/bug.h:5,
from include/linux/mmdebug.h:5,
from include/linux/mm.h:9,
from mm/page_alloc.c:18:
mm/page_alloc.c: In function 'deferred_grow_zone':
mm/page_alloc.c:1624:52: error: 'PAGES_PER_SECTION' undeclared (first use in this function); did you mean 'USEC_PER_SEC'?
unsigned long nr_pages_needed = ALIGN(1 << order, PAGES_PER_SECTION);
^
include/uapi/linux/kernel.h:11:47: note: in definition of macro '__ALIGN_KERNEL_MASK'
#define __ALIGN_KERNEL_MASK(x, mask) (((x) + (mask)) & ~(mask))
^~~~
include/linux/kernel.h:58:22: note: in expansion of macro '__ALIGN_KERNEL'
#define ALIGN(x, a) __ALIGN_KERNEL((x), (a))
^~~~~~~~~~~~~~
mm/page_alloc.c:1624:34: note: in expansion of macro 'ALIGN'
unsigned long nr_pages_needed = ALIGN(1 << order, PAGES_PER_SECTION);
^~~~~
In file included from include/asm-generic/bug.h:18:0,
from arch/arc/include/asm/bug.h:32,
from include/linux/bug.h:5,
from include/linux/mmdebug.h:5,
from include/linux/mm.h:9,
from mm/page_alloc.c:18:
mm/page_alloc.c: In function 'free_area_init_node':
mm/page_alloc.c:6379:50: error: 'PAGES_PER_SECTION' undeclared (first use in this function); did you mean 'USEC_PER_SEC'?
pgdat->static_init_pgcnt = min_t(unsigned long, PAGES_PER_SECTION,
^
include/linux/kernel.h:812:22: note: in definition of macro '__typecheck'
(!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
^
include/linux/kernel.h:836:24: note: in expansion of macro '__safe_cmp'
__builtin_choose_expr(__safe_cmp(x, y), \
^~~~~~~~~~
include/linux/kernel.h:904:27: note: in expansion of macro '__careful_cmp'
#define min_t(type, x, y) __careful_cmp((type)(x), (type)(y), <)
^~~~~~~~~~~~~
mm/page_alloc.c:6379:29: note: in expansion of macro 'min_t'
pgdat->static_init_pgcnt = min_t(unsigned long, PAGES_PER_SECTION,
^~~~~
include/linux/kernel.h:836:2: error: first argument to '__builtin_choose_expr' not a constant
__builtin_choose_expr(__safe_cmp(x, y), \
^
include/linux/kernel.h:904:27: note: in expansion of macro '__careful_cmp'
#define min_t(type, x, y) __careful_cmp((type)(x), (type)(y), <)
^~~~~~~~~~~~~
mm/page_alloc.c:6379:29: note: in expansion of macro 'min_t'
pgdat->static_init_pgcnt = min_t(unsigned long, PAGES_PER_SECTION,
^~~~~ scripts/Makefile.build:317: recipe for target 'mm/page_alloc.o' failed
Let's make the DEFERRED_STRUCT_PAGE_INIT explicitly depend on SPARSEMEM as the systems that support DISCONTIGMEM do not seem to have that huge amounts of memory that would make DEFERRED_STRUCT_PAGE_INIT relevant.
--- mm/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
mm/Kconfig +1 -1
@@ -635,7 +635,7 @@ config DEFERRED_STRUCT_PAGE_INIT bool "Defer initialisation of struct pages to kthreads" default n depends on NO_BOOTMEM - depends on !FLATMEM + depends on SPARSEMEM depends on !NEED_PER_CPU_KM help Ordinarily all struct pages are initialised during early boot in a
Greg Kroah-Hartman
maintainerPATCH 61/197 fs/dcache.c: fix kmemcheck splat at take_dentry_name_snapshot()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tetsuo Handa <[email protected]>
[ Upstream commit 6cd00a01f0c1ae6a852b09c59b8dd55cc6c35d1d ]
Since only dentry->d_name.len + 1 bytes out of DNAME_INLINE_LEN bytes are initialized at __d_alloc(), we can't copy the whole size unconditionally.
WARNING: kmemcheck: Caught 32-bit read from uninitialized memory (ffff8fa27465ac50) 636f6e66696766732e746d70000000000010000000000000020000000188ffff
i i i i i i i i i i i i i u u u u u u u u u u i i i i i u u u u
^ RIP: 0010:take_dentry_name_snapshot+0x28/0x50 RSP: 0018:ffffa83000f5bdf8 EFLAGS: 00010246 RAX: 0000000000000020 RBX: ffff8fa274b20550 RCX: 0000000000000002 RDX: ffffa83000f5be40 RSI: ffff8fa27465ac50 RDI: ffffa83000f5be60 RBP: ffffa83000f5bdf8 R08: ffffa83000f5be48 R09: 0000000000000001 R10: ffff8fa27465ac00 R11: ffff8fa27465acc0 R12: ffff8fa27465ac00 R13: ffff8fa27465acc0 R14: 0000000000000000 R15: 0000000000000000 FS: 00007f79737ac8c0(0000) GS:ffffffff8fc30000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffff8fa274c0b000 CR3: 0000000134aa7002 CR4: 00000000000606f0 take_dentry_name_snapshot+0x28/0x50 vfs_rename+0x128/0x870 SyS_rename+0x3b2/0x3d0 entry_SYSCALL_64_fastpath+0x1a/0xa4 0xffffffffffffffff
--- fs/dcache.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
fs/dcache.c +2 -1
@@ -292,7 +292,8 @@ void take_dentry_name_snapshot(struct naspin_unlock (&dentry->d_lock); name->name = p->name; }else { -memcpy (name->inline_name, dentry->d_iname,DNAME_INLINE_LEN ); +memcpy (name->inline_name, dentry->d_iname, + dentry->d_name.len +1 );spin_unlock (&dentry->d_lock); name->name = name->inline_name; }
Greg Kroah-Hartman
maintainerPATCH 62/197 platform/x86: intel_punit_ipc: fix build errors
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Randy Dunlap <[email protected]>
[ Upstream commit 340fd4cff43f18bace9358d4decdc9b6ed0715be ]
Fix build errors by #including <linux/io.h>.
../drivers/platform/x86/intel_punit_ipc.c: In function 'ipc_read_status':../drivers/platform/x86/intel_punit_ipc.c:55:2: error: implicit declaration of function 'readl' [-Werror=implicit-function-declaration]
return readl(ipcdev->base[type][BASE_IFACE]);../drivers/platform/x86/intel_punit_ipc.c: In function 'ipc_write_cmd':../drivers/platform/x86/intel_punit_ipc.c:60:2: error: implicit declaration of function 'writel' [-Werror=implicit-function-declaration]
writel(cmd, ipcdev->base[type][BASE_IFACE]);
--- drivers/platform/x86/intel_punit_ipc.c | 1 + 1 file changed, 1 insertion(+)
drivers/platform/x86/intel_punit_ipc.c +1 -0
@@ -17,6 +17,7 @@#include <linux/bitops.h> #include <linux/device.h> #include <linux/interrupt.h> +#include <linux/io.h> #include <linux/platform_device.h> #include <asm/intel_punit_ipc.h>
Greg Kroah-Hartman
maintainerPATCH 63/197 bpf, sockmap: fix map elem deletion race with smap_stop_sock
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Borkmann <[email protected]>
[ Upstream commit 166ab6f0a0702fdd4d865ad5090bf3094ed83428 ]
The smap_start_sock() and smap_stop_sock() are each protected under the sock->sk_callback_lock from their call-sites except in the case of sock_map_delete_elem() where we drop the old socket from the map slot. This is racy because the same sock could be part of multiple sock maps, so we run smap_stop_sock() in parallel, and given at that point psock->strp_enabled might be true on both CPUs, we might for example wrongly restore the sk->sk_data_ready / sk->sk_write_space. Therefore, hold the sock->sk_callback_lock as well on delete. Looks like 2f857d04601a ("bpf: sockmap, remove STRPARSER map_flags and add multi-map support") had this right, but later on e9db4ef6bf4c ("bpf:
sockhash fix omitted bucket lock in sock_close") removed it again from delete leaving this smap_stop_sock() instance unprotected.
--- kernel/bpf/sockmap.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
kernel/bpf/sockmap.c +4 -1
@@ -1784,8 +1784,11 @@ static int sock_map_delete_elem(struct bif (!psock) goto out; -if (psock->bpf_parse) +if (psock->bpf_parse) { +write_lock_bh (&sock->sk_callback_lock);smap_stop_sock (psock, sock); +write_unlock_bh (&sock->sk_callback_lock); + }smap_list_map_remove (psock, &stab->sock_map[k]);smap_release_sock (psock, sock); out:
Greg Kroah-Hartman
maintainerPATCH 64/197 tcp, ulp: fix leftover icsk_ulp_ops preventing sock from reattach
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Borkmann <[email protected]>
[ Upstream commit 90545cdc3f2b2ea700e24335610cd181e73756da ]
I found that in BPF sockmap programs once we either delete a socket from the map or we updated a map slot and the old socket was purged from the map that these socket can never get reattached into a map even though their related psock has been dropped entirely at that point.
Reason is that tcp_cleanup_ulp() leaves the old icsk->icsk_ulp_ops intact, so that on the next tcp_set_ulp_id() the kernel returns an -EEXIST thinking there is still some active ULP attached.
BPF sockmap is the only one that has this issue as the other user, kTLS, only calls tcp_cleanup_ulp() from tcp_v4_destroy_sock() whereas sockmap semantics allow dropping the socket from the map with all related psock state being cleaned up.
--- net/ipv4/tcp_ulp.c | 2 ++ 1 file changed, 2 insertions(+)
net/ipv4/tcp_ulp.c +2 -0
@@ -129,6 +129,8 @@ void tcp_cleanup_ulp(struct sock *sk)if (icsk->icsk_ulp_ops->release) icsk->icsk_ulp_ops->release (sk);module_put (icsk->icsk_ulp_ops->owner); + + icsk->icsk_ulp_ops =NULL ; }/* Change upper layer protocol for socket */
Greg Kroah-Hartman
maintainerPATCH 65/197 bpf, sockmap: fix sock_map_ctx_update_elem race with exist/noexist
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Borkmann <[email protected]>
[ Upstream commit 585f5a6252ee43ec8feeee07387e3fcc7e8bb292 ]
The current code in sock_map_ctx_update_elem() allows for BPF_EXIST and BPF_NOEXIST map update flags. While on array-like maps this approach is rather uncommon, e.g. bpf_fd_array_map_update_elem() and others enforce map update flags to be BPF_ANY such that xchg() can be used directly, the current implementation in sock map does not guarantee that such operation with BPF_EXIST / BPF_NOEXIST is atomic.
The initial test does a READ_ONCE(stab->sock_map[i]) to fetch the socket from the slot which is then tested for NULL / non-NULL. However later after __sock_map_ctx_update_elem(), the actual update is done through osock = xchg(&stab->sock_map[i], sock). Problem is that in the meantime a different CPU could have updated / deleted a socket on that specific slot and thus flag contraints won't hold anymore.
I've been thinking whether best would be to just break UAPI and do an enforcement of BPF_ANY to check if someone actually complains, however trouble is that already in BPF kselftest we use BPF_NOEXIST for the map update, and therefore it might have been copied into applications already. The fix to keep the current behavior intact would be to add a map lock similar to the sock hash bucket lock only for covering the whole map.
--- kernel/bpf/sockmap.c | 106 +++++++++++++++++++++++++++------------------------ 1 file changed, 57 insertions(+), 49 deletions(-)
kernel/bpf/sockmap.c +57 -49
@@ -58,6 +58,7 @@ struct bpf_stab {struct bpf_map map;struct sock **sock_map;struct bpf_sock_progs progs; +raw_spinlock_t lock; };struct bucket {
@@ -89,9 +90,9 @@ enum smap_psock_state {struct smap_psock_map_entry {struct list_head list; +struct bpf_map *map;struct sock **entry;struct htab_elem __rcu *hash_link; -struct bpf_htab __rcu *htab; };struct smap_psock {
@@ -343,13 +344,18 @@ static void bpf_tcp_close(struct sock *s e =psock_map_pop (sk, psock);while (e) {if (e->entry) { - osk =cmpxchg (e->entry, sk,NULL ); + struct bpf_stab *stab = container_of(e->map, struct bpf_stab, map); + +raw_spin_lock_bh (&stab->lock); + osk = *e->entry;if (osk == sk) { + *e->entry =NULL ;smap_release_sock (psock, sk); } +raw_spin_unlock_bh (&stab->lock); }else {struct htab_elem *link =rcu_dereference (e->hash_link); -struct bpf_htab *htab =rcu_dereference (e->htab); + struct bpf_htab *htab = container_of(e->map, struct bpf_htab, map);struct hlist_head *head;struct htab_elem *l;struct bucket *b;
@@ -1644,6 +1650,7 @@ static struct bpf_map *sock_map_alloc(unreturn ERR_PTR (-ENOMEM );bpf_map_init_from_attr (&stab->map, attr); +raw_spin_lock_init (&stab->lock);/* make sure page count doesn't overflow */ cost = (u64 ) stab->map.max_entries *sizeof (struct sock *);
@@ -1714,14 +1721,15 @@ static void sock_map_free(struct bpf_map* and a grace period expire to ensure psock is really safe to remove. */ rcu_read_lock (); +raw_spin_lock_bh (&stab->lock);for (i =0 ; i < stab->map.max_entries; i++) {struct smap_psock *psock;struct sock *sock; - sock =xchg (&stab->sock_map[i],NULL ); + sock = stab->sock_map[i];if (!sock)continue ; - + stab->sock_map[i] =NULL ; psock =smap_psock_sk (sock);/* This check handles a racing sock event that can get the * sk_callback_lock before this case but after xchg happens
@@ -1733,6 +1741,7 @@ static void sock_map_free(struct bpf_mapsmap_release_sock (psock, sock); } } +raw_spin_unlock_bh (&stab->lock);rcu_read_unlock ();sock_map_remove_complete (stab);
@@ -1776,14 +1785,16 @@ static int sock_map_delete_elem(struct bif (k >= map->max_entries)return -EINVAL ; - sock =xchg (&stab->sock_map[k],NULL ); +raw_spin_lock_bh (&stab->lock); + sock = stab->sock_map[k]; + stab->sock_map[k] =NULL ; +raw_spin_unlock_bh (&stab->lock);if (!sock)return -EINVAL ; psock =smap_psock_sk (sock);if (!psock) - goto out; - +return 0 ;if (psock->bpf_parse) {write_lock_bh (&sock->sk_callback_lock);smap_stop_sock (psock, sock);
@@ -1791,7 +1802,6 @@ static int sock_map_delete_elem(struct b }smap_list_map_remove (psock, &stab->sock_map[k]);smap_release_sock (psock, sock); -out:return 0 ; }
@@ -1827,11 +1837,9 @@ out:static int __sock_map_ctx_update_elem (struct bpf_map *map,struct bpf_sock_progs *progs,struct sock *sock, -struct sock **map_link,void *key) {struct bpf_prog *verdict, *parse, *tx_msg; -struct smap_psock_map_entry *e =NULL ;struct smap_psock *psock;bool new = false;int err =0 ;
@@ -1904,14 +1912,6 @@ static int __sock_map_ctx_update_elem(st new = true; } -if (map_link) { - e =kzalloc (sizeof (*e),GFP_ATOMIC | __GFP_NOWARN); -if (!e) { - err = -ENOMEM ; - goto out_free; - } - } -/* 3. At this point we have a reference to a valid psock that is * running. Attach any BPF programs needed. */
@@ -1933,17 +1933,6 @@ static int __sock_map_ctx_update_elem(stwrite_unlock_bh (&sock->sk_callback_lock); } -/* 4. Place psock in sockmap for use and stop any programs on -* the old sock assuming its not the same sock we are replacing -* it with. Because we can only have a single set of programs if -* old_sock has a strp we can stop it. -*/ -if (map_link) { - e->entry = map_link; -spin_lock_bh (&psock->maps_lock); -list_add_tail (&e->list, &psock->maps); -spin_unlock_bh (&psock->maps_lock); - }return err; out_free:smap_release_sock (psock, sock);
@@ -1954,7 +1943,6 @@ out_progs: }if (tx_msg)bpf_prog_put (tx_msg); -kfree (e);return err; }
@@ -1964,36 +1952,57 @@ static int sock_map_ctx_update_elem(stru { struct bpf_stab *stab = container_of(map, struct bpf_stab, map);struct bpf_sock_progs *progs = &stab->progs; -struct sock *osock, *sock; +struct sock *osock, *sock = skops->sk; +struct smap_psock_map_entry *e; +struct smap_psock *psock;u32 i = *(u32 *)key;int err;if (unlikely (flags >BPF_EXIST ))return -EINVAL ; -if (unlikely (i >= stab->map.max_entries))return -E2BIG ; - sock =READ_ONCE (stab->sock_map[i]); -if (flags ==BPF_EXIST && !sock) -return -ENOENT ; -else if (flags ==BPF_NOEXIST && sock) -return -EEXIST ; + e =kzalloc (sizeof (*e),GFP_ATOMIC | __GFP_NOWARN); +if (!e) +return -ENOMEM ; - sock = skops->sk; - err =__sock_map_ctx_update_elem (map, progs, sock, &stab->sock_map[i], - key); + err =__sock_map_ctx_update_elem (map, progs, sock, key);if (err) goto out; - osock =xchg (&stab->sock_map[i], sock); -if (osock) { -struct smap_psock *opsock =smap_psock_sk (osock); +/* psock guaranteed to be present. */ + psock =smap_psock_sk (sock); +raw_spin_lock_bh (&stab->lock); + osock = stab->sock_map[i]; +if (osock && flags ==BPF_NOEXIST ) { + err = -EEXIST ; + goto out_unlock; + } +if (!osock && flags ==BPF_EXIST ) { + err = -ENOENT ; + goto out_unlock; + } + + e->entry = &stab->sock_map[i]; + e->map = map; +spin_lock_bh (&psock->maps_lock); +list_add_tail (&e->list, &psock->maps); +spin_unlock_bh (&psock->maps_lock); -smap_list_map_remove (opsock, &stab->sock_map[i]); -smap_release_sock (opsock, osock); + stab->sock_map[i] = sock; +if (osock) { + psock =smap_psock_sk (osock); +smap_list_map_remove (psock, &stab->sock_map[i]); +smap_release_sock (psock, osock); } +raw_spin_unlock_bh (&stab->lock); +return 0 ; +out_unlock: +smap_release_sock (psock, sock); +raw_spin_unlock_bh (&stab->lock); out: +kfree (e);return err; }
@@ -2356,7 +2365,7 @@ static int sock_hash_ctx_update_elem(str b =__select_bucket (htab, hash); head = &b->head; - err =__sock_map_ctx_update_elem (map, progs, sock,NULL , key); + err =__sock_map_ctx_update_elem (map, progs, sock, key);if (err) goto err;
@@ -2382,8 +2391,7 @@ static int sock_hash_ctx_update_elem(str }rcu_assign_pointer (e->hash_link, l_new); -rcu_assign_pointer (e->htab, - container_of(map, struct bpf_htab, map)); + e->map = map;spin_lock_bh (&psock->maps_lock);list_add_tail (&e->list, &psock->maps);spin_unlock_bh (&psock->maps_lock);
Greg Kroah-Hartman
maintainerPATCH 66/197 net/xdp: Fix suspicious RCU usage warning
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tariq Toukan <[email protected]>
[ Upstream commit 21b172ee11b6ec260bd7e6a27b11a8a8d392fce5 ]
Fix the warning below by calling rhashtable_lookup_fast. Also, make some code movements for better quality and human readability.
[ 342.450870] WARNING: suspicious RCU usage
[ 342.455856] 4.18.0-rc2+ #17 Tainted: G O
[ 342.462210] -----------------------------
[ 342.467202] ./include/linux/rhashtable.h:481 suspicious rcu_dereference_check() usage!
[ 342.476568]
[ 342.476568] other info that might help us debug this:
[ 342.476568]
[ 342.486978]
[ 342.486978] rcu_scheduler_active = 2, debug_locks = 1
[ 342.495211] 4 locks held by modprobe/3934:
[ 342.500265] #0: 00000000e23116b2 (mlx5_intf_mutex){+.+.}, at:
mlx5_unregister_interface+0x18/0x90 [mlx5_core]
[ 342.511953] #1: 00000000ca16db96 (rtnl_mutex){+.+.}, at: unregister_netdev+0xe/0x20
[ 342.521109] #2: 00000000a46e2c4b (&priv->state_lock){+.+.}, at: mlx5e_close+0x29/0x60
[mlx5_core]
[ 342.531642] #3: 0000000060c5bde3 (mem_id_lock){+.+.}, at: xdp_rxq_info_unreg+0x93/0x6b0
[ 342.541206]
[ 342.541206] stack backtrace:
[ 342.547075] CPU: 12 PID: 3934 Comm: modprobe Tainted: G O 4.18.0-rc2+ #17
[ 342.556621] Hardware name: Dell Inc. PowerEdge R730/0H21J3, BIOS 1.5.4 10/002/2015
[ 342.565606] Call Trace:
[ 342.568861] dump_stack+0x78/0xb3
[ 342.573086] xdp_rxq_info_unreg+0x3f5/0x6b0
[ 342.578285] ? __call_rcu+0x220/0x300
[ 342.582911] mlx5e_free_rq+0x38/0xc0 [mlx5_core]
[ 342.588602] mlx5e_close_channel+0x20/0x120 [mlx5_core]
[ 342.594976] mlx5e_close_channels+0x26/0x40 [mlx5_core]
[ 342.601345] mlx5e_close_locked+0x44/0x50 [mlx5_core]
[ 342.607519] mlx5e_close+0x42/0x60 [mlx5_core]
[ 342.613005] __dev_close_many+0xb1/0x120
[ 342.617911] dev_close_many+0xa2/0x170
[ 342.622622] rollback_registered_many+0x148/0x460
[ 342.628401] ? __lock_acquire+0x48d/0x11b0
[ 342.633498] ? unregister_netdev+0xe/0x20
[ 342.638495] rollback_registered+0x56/0x90
[ 342.643588] unregister_netdevice_queue+0x7e/0x100
[ 342.649461] unregister_netdev+0x18/0x20
[ 342.654362] mlx5e_remove+0x2a/0x50 [mlx5_core]
[ 342.659944] mlx5_remove_device+0xe5/0x110 [mlx5_core]
[ 342.666208] mlx5_unregister_interface+0x39/0x90 [mlx5_core]
[ 342.673038] cleanup+0x5/0xbfc [mlx5_core]
[ 342.678094] __x64_sys_delete_module+0x16b/0x240
[ 342.683725] ? do_syscall_64+0x1c/0x210
[ 342.688476] do_syscall_64+0x5a/0x210
[ 342.693025] entry_SYSCALL_64_after_hwframe+0x49/0xbe--- net/core/xdp.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-)
net/core/xdp.c +3 -11
@@ -95,23 +95,15 @@ static void __xdp_rxq_info_unreg_mem_mod {struct xdp_mem_allocator *xa;int id = xdp_rxq->mem.id; -int err;if (id ==0 )return ;mutex_lock (&mem_id_lock); - xa =rhashtable_lookup (mem_id_ht, &id, mem_id_rht_params); -if (!xa) { -mutex_unlock (&mem_id_lock); -return ; - } - - err =rhashtable_remove_fast (mem_id_ht, &xa->node, mem_id_rht_params); -WARN_ON (err); - -call_rcu (&xa->rcu, __xdp_mem_allocator_rcu_free); + xa =rhashtable_lookup_fast (mem_id_ht, &id, mem_id_rht_params); +if (xa && !rhashtable_remove_fast (mem_id_ht, &xa->node, mem_id_rht_params)) +call_rcu (&xa->rcu, __xdp_mem_allocator_rcu_free);mutex_unlock (&mem_id_lock); }
Greg Kroah-Hartman
maintainerPATCH 67/197 bpf, sockmap: fix leakage of smap_psock_map_entry
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Borkmann <[email protected]>
[ Upstream commit d40b0116c94bd8fc2b63aae35ce8e66bb53bba42 ]
While working on sockmap I noticed that we do not always kfree the struct smap_psock_map_entry list elements which track psocks attached to maps. In the case of sock_hash_ctx_update_elem(), these map entries are allocated outside of __sock_map_ctx_update_elem() with their linkage to the socket hash table filled. In the case of sock array, the map entries are allocated inside of __sock_map_ctx_update_elem() and added with their linkage to the psock->maps. Both additions are under psock->maps_lock each.
Now, we drop these elements from their psock->maps list in a few
occasions: i) in sock array via smap_list_map_remove() when an entry is either deleted from the map from user space, or updated via user space or BPF program where we drop the old socket at that map slot, or the sock array is freed via sock_map_free() and drops all its elements; ii) for sock hash via smap_list_hash_remove() in exactly the same occasions as just described for sock array; iii) in the bpf_tcp_close() where we remove the elements from the list via psock_map_pop() and iterate over them dropping themselves from either sock array or sock hash; and last but not least iv) once again in smap_gc_work() which is a callback for deferring the work once the psock refcount hit zero and thus the socket is being destroyed.
Problem is that the only case where we kfree() the list entry is in case iv), which at that point should have an empty list in normal cases. So in cases from i) to iii) we unlink the elements without freeing where they go out of reach from us. Hence fix is to properly kfree() them as well to stop the leakage. Given these are all handled under psock->maps_lock there is no need for deferred RCU freeing.
I later also ran with kmemleak detector and it confirmed the finding as well where in the state before the fix the object goes unreferenced while after the patch no kmemleak report related to BPF showed up.
[...]
unreferenced object 0xffff880378eadae0 (size 64):
comm "test_sockmap", pid 2225, jiffies 4294720701 (age 43.504s)
hex dump (first 32 bytes):
00 01 00 00 00 00 ad de 00 02 00 00 00 00 ad de ................
50 4d 75 5d 03 88 ff ff 00 00 00 00 00 00 00 00 PMu]............
backtrace:
[<000000005225ac3c>] sock_map_ctx_update_elem.isra.21+0xd8/0x210
[<0000000045dd6d3c>] bpf_sock_map_update+0x29/0x60
[<00000000877723aa>] ___bpf_prog_run+0x1e1f/0x4960
[<000000002ef89e83>] 0xffffffffffffffff
unreferenced object 0xffff880378ead240 (size 64):
comm "test_sockmap", pid 2225, jiffies 4294720701 (age 43.504s)
hex dump (first 32 bytes):
00 01 00 00 00 00 ad de 00 02 00 00 00 00 ad de ................
00 44 75 5d 03 88 ff ff 00 00 00 00 00 00 00 00 .Du]............
backtrace:
[<000000005225ac3c>] sock_map_ctx_update_elem.isra.21+0xd8/0x210
[<0000000030e37a3a>] sock_map_update_elem+0x125/0x240
[<000000002e5ce36e>] map_update_elem+0x4eb/0x7b0
[<00000000db453cc9>] __x64_sys_bpf+0x1f9/0x360
[<0000000000763660>] do_syscall_64+0x9a/0x300
[<00000000422a2bb2>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[<000000002ef89e83>] 0xffffffffffffffff[...]
--- kernel/bpf/sockmap.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
kernel/bpf/sockmap.c +7 -2
@@ -376,6 +376,7 @@ static void bpf_tcp_close(struct sock *s }raw_spin_unlock_bh (&b->lock); } +kfree (e); e =psock_map_pop (sk, psock); }rcu_read_unlock ();
@@ -1685,8 +1686,10 @@ static void smap_list_map_remove(structspin_lock_bh (&psock->maps_lock);list_for_each_entry_safe (e, tmp, &psock->maps, list) { -if (e->entry == entry) +if (e->entry == entry) {list_del (&e->list); +kfree (e); + } }spin_unlock_bh (&psock->maps_lock); }
@@ -1700,8 +1703,10 @@ static void smap_list_hash_remove(structlist_for_each_entry_safe (e, tmp, &psock->maps, list) {struct htab_elem *c =rcu_dereference (e->hash_link); -if (c == hash_link) +if (c == hash_link) {list_del (&e->list); +kfree (e); + } }spin_unlock_bh (&psock->maps_lock); }
Greg Kroah-Hartman
maintainerPATCH 68/197 samples/bpf: all XDP samples should unload xdp/bpf prog on SIGTERM
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jesper Dangaard Brouer <[email protected]>
[ Upstream commit 817b89beb9d8876450fcde9155e17425c329569d ]
It is common XDP practice to unload/deattach the XDP bpf program, when the XDP sample program is Ctrl-C interrupted (SIGINT) or killed (SIGTERM).
The samples/bpf programs xdp_redirect_cpu and xdp_rxq_info, forgot to trap signal SIGTERM (which is the default signal used by the kill command).
This was discovered by Red Hat QA, which automated scripts depend on killing the XDP sample program after a timeout period.
--- samples/bpf/xdp_redirect_cpu_user.c | 3 ++- samples/bpf/xdp_rxq_info_user.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-)
samples/bpf/xdp_redirect_cpu_user.c +2 -1
@@ -679,8 +679,9 @@ int main(int argc, char **argv)return EXIT_FAIL_OPTION ; } -/* Remove XDP program when program is interrupted */ +/* Remove XDP program when program is interrupted or killed */ signal (SIGINT , int_exit); +signal (SIGTERM , int_exit);if (bpf_set_link_xdp_fd (ifindex, prog_fd[prog_num], xdp_flags) <0 ) {fprintf (stderr,"link set xdp fd failed\n" );
samples/bpf/xdp_rxq_info_user.c +2 -1
@@ -534,8 +534,9 @@ int main(int argc, char **argv)exit (EXIT_FAIL_BPF ); } -/* Remove XDP program when program is interrupted */ +/* Remove XDP program when program is interrupted or killed */ signal (SIGINT , int_exit); +signal (SIGTERM , int_exit);if (bpf_set_link_xdp_fd (ifindex, prog_fd, xdp_flags) <0 ) {fprintf (stderr,"link set xdp fd failed\n" );
Greg Kroah-Hartman
maintainerPATCH 69/197 netfilter: ip6t_rpfilter: set F_IFACE for linklocal addresses
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Westphal <[email protected]>
[ Upstream commit da786717e0894886301ed2536843c13f9e8fd53e ]
Roman reports that DHCPv6 client no longer sees replies from server due to
ip6tables -t raw -A PREROUTING -m rpfilter --invert -j DROP
rule. We need to set the F_IFACE flag for linklocal addresses, they are scoped per-device.
--- net/ipv6/netfilter/ip6t_rpfilter.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)
net/ipv6/netfilter/ip6t_rpfilter.c +11 -1
@@ -26,6 +26,12 @@ static bool rpfilter_addr_unicast(constreturn addr_type &IPV6_ADDR_UNICAST ; } +static bool rpfilter_addr_linklocal (const struct in6_addr *addr) +{ +int addr_type =ipv6_addr_type (addr); +return addr_type &IPV6_ADDR_LINKLOCAL ; +} +static bool rpfilter_lookup_reverse6 (struct net *net,const struct sk_buff *skb,const struct net_device *dev,u8 flags) {
@@ -48,7 +54,11 @@ static bool rpfilter_lookup_reverse6(str } fl6.flowi6_mark = flags &XT_RPFILTER_VALID_MARK ? skb->mark :0 ; -if ((flags &XT_RPFILTER_LOOSE ) ==0 ) + +if (rpfilter_addr_linklocal (&iph->saddr)) { + lookup_flags |=RT6_LOOKUP_F_IFACE ; + fl6.flowi6_oif = dev->ifindex; + }else if ((flags &XT_RPFILTER_LOOSE ) ==0 ) fl6.flowi6_oif = dev->ifindex; rt = (void *)ip6_route_lookup (net, &fl6, skb, lookup_flags);
Greg Kroah-Hartman
maintainerPATCH 70/197 s390/kdump: Fix memleak in nt_vmcoreinfo
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Philipp Rudo <[email protected]>
[ Upstream commit 2d2e7075b87181ed0c675e4936e20bdadba02e1f ]
The vmcoreinfo of a crashed system is potentially fragmented. Thus the crash kernel has an intermediate step where the vmcoreinfo is copied into a temporary, continuous buffer in the crash kernel memory. This temporary buffer is never freed. Free it now to prevent the memleak.
While at it replace all occurrences of "VMCOREINFO" by its corresponding macro to prevent potential renaming issues.
--- arch/s390/kernel/crash_dump.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)
arch/s390/kernel/crash_dump.c +12 -5
@@ -404,11 +404,13 @@ static void *get_vmcoreinfo_old(unsignedif (copy_oldmem_kernel (nt_name, addr +sizeof (note),sizeof (nt_name) -1 ))return NULL ; -if (strcmp (nt_name,"VMCOREINFO" ) !=0 ) +if (strcmp (nt_name,VMCOREINFO_NOTE_NAME ) !=0 )return NULL ; vmcoreinfo =kzalloc_panic (note.n_descsz); -if (copy_oldmem_kernel (vmcoreinfo, addr +24 , note.n_descsz)) +if (copy_oldmem_kernel (vmcoreinfo, addr +24 , note.n_descsz)) { +kfree (vmcoreinfo);return NULL ; + } *size = note.n_descsz;return vmcoreinfo; }
@@ -418,15 +420,20 @@ static void *get_vmcoreinfo_old(unsigned*/ static void *nt_vmcoreinfo (void *ptr) { +const char *name =VMCOREINFO_NOTE_NAME ;unsigned long size;void *vmcoreinfo; vmcoreinfo =os_info_old_entry (OS_INFO_VMCOREINFO , &size); -if (!vmcoreinfo) - vmcoreinfo =get_vmcoreinfo_old (&size); +if (vmcoreinfo) +return nt_init_name (ptr,0 , vmcoreinfo, size, name); + + vmcoreinfo =get_vmcoreinfo_old (&size);if (!vmcoreinfo)return ptr; -return nt_init_name (ptr,0 , vmcoreinfo, size,"VMCOREINFO" ); + ptr =nt_init_name (ptr,0 , vmcoreinfo, size, name); +kfree (vmcoreinfo); +return ptr; }/*
Greg Kroah-Hartman
maintainerPATCH 71/197 ipvs: fix race between ip_vs_conn_new() and ip_vs_del_dest()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tan Hu <[email protected]>
[ Upstream commit a53b42c11815d2357e31a9403ae3950517525894 ]
We came across infinite loop in ipvs when using ipvs in docker env.
When ipvs receives new packets and cannot find an ipvs connection, it will create a new connection, then if the dest is unavailable (i.e. IP_VS_DEST_F_AVAILABLE), the packet will be dropped sliently.
But if the dropped packet is the first packet of this connection, the connection control timer never has a chance to start and the ipvs connection cannot be released. This will lead to memory leak, or infinite loop in cleanup_net() when net namespace is released like this:
ip_vs_conn_net_cleanup at ffffffffa0a9f31a [ip_vs] __ip_vs_cleanup at ffffffffa0a9f60a [ip_vs] ops_exit_list at ffffffff81567a49 cleanup_net at ffffffff81568b40 process_one_work at ffffffff810a851b worker_thread at ffffffff810a9356
kthread at ffffffff810b0b6f
ret_from_fork at ffffffff81697a18
race condition:
CPU1 CPU2
ip_vs_in()
ip_vs_conn_new()
ip_vs_del_dest()
__ip_vs_unlink_dest()
~IP_VS_DEST_F_AVAILABLE
cp->dest && !IP_VS_DEST_F_AVAILABLE
__ip_vs_conn_put
...
cleanup_net ---> infinite loopingFix this by checking whether the timer already started.
--- net/netfilter/ipvs/ip_vs_core.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-)
net/netfilter/ipvs/ip_vs_core.c +11 -4
@@ -1972,13 +1972,20 @@ ip_vs_in(struct netns_ipvs *ipvs, unsignif (cp->dest && !(cp->dest->flags &IP_VS_DEST_F_AVAILABLE )) {/* the destination server is not available */ -if (sysctl_expire_nodest_conn (ipvs)) { +__u32 flags = cp->flags; + +/* when timer already started, silently drop the packet.*/ +if (timer_pending (&cp->timer)) +__ip_vs_conn_put (cp); +else +ip_vs_conn_put (cp); + +if (sysctl_expire_nodest_conn (ipvs) && + !(flags &IP_VS_CONN_F_ONE_PACKET )) {/* try to expire the connection immediately */ ip_vs_conn_expire_now (cp); } -/* don't restart its timer, and silently -drop the packet. */ -__ip_vs_conn_put (cp); +return NF_DROP ; }
Greg Kroah-Hartman
maintainerPATCH 72/197 mfd: sm501: Set coherent_dma_mask when creating subdevices
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Guenter Roeck <[email protected]>
[ Upstream commit 2f606da78230f09cf1a71fde6ee91d0c710fa2b2 ]
Instantiating the sm501 OHCI subdevice results in a kernel warning.
sm501-usb sm501-usb: SM501 OHCI
sm501-usb sm501-usb: new USB bus registered, assigned bus number 1
WARNING: CPU: 0 PID: 1 at ./include/linux/dma-mapping.h:516 ohci_init+0x194/0x2d8 Modules linked in: CPU: 0 PID: 1 Comm: swapper Tainted: G W
4.18.0-rc7-00178-g0b5b1f9a78b5 #1
PC is at ohci_init+0x194/0x2d8
PR is at ohci_init+0x168/0x2d8
PC : 8c27844c SP : 8f81dd94 SR : 40008001
TEA : 29613060
R0 : 00000000 R1 : 00000000 R2 : 00000000 R3 : 00000202 R4 : 8fa98b88 R5 : 8c277e68 R6 : 00000000 R7 : 00000000 R8 : 8f965814 R9 : 8c388100 R10 : 8fa98800 R11 : 8fa98928 R12 : 8c48302c R13 : 8fa98920 R14 : 8c48302c
MACH: 00000096 MACL: 0000017c GBR : 00000000 PR : 8c278420
Call trace: [<(ptrval)>] usb_add_hcd+0x1e8/0x6ec [<(ptrval)>] _dev_info+0x0/0x54 [<(ptrval)>] arch_local_save_flags+0x0/0x8 [<(ptrval)>] arch_local_irq_restore+0x0/0x24 [<(ptrval)>] ohci_hcd_sm501_drv_probe+0x114/0x2d8
...
Initialize coherent_dma_mask when creating SM501 subdevices to fix the problem.
--- drivers/mfd/sm501.c | 1 + 1 file changed, 1 insertion(+)
drivers/mfd/sm501.c +1 -0
@@ -715,6 +715,7 @@ sm501_create_subdev(struct sm501_devdata smdev->pdev.name = name; smdev->pdev.id = sm->pdev_id; smdev->pdev.dev.parent = sm->dev; + smdev->pdev.dev.coherent_dma_mask =0xffffffff ;if (res_count) { smdev->pdev.resource = (struct resource *)(smdev+1 );
Greg Kroah-Hartman
maintainerPATCH 73/197 netfilter: x_tables: do not fail xt_alloc_table_info too easilly
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michal Hocko <[email protected]>
[ Upstream commit a148ce15375fc664ad64762c751c0c2aecb2cafe ]
eacd86ca3b03 ("net/netfilter/x_tables.c: use kvmalloc()
in xt_alloc_table_info()") has unintentionally fortified xt_alloc_table_info allocation when __GFP_RETRY has been dropped from the vmalloc fallback. Later on there was a syzbot report that this can lead to OOM killer invocations when tables are too large and 0537250fdc6c ("netfilter: x_tables: make allocation less aggressive") has been merged to restore the original behavior. Georgi Nikolov however noticed that he is not able to install his iptables anymore so this can be seen as a regression.
The primary argument for 0537250fdc6c was that this allocation path shouldn't really trigger the OOM killer and kill innocent tasks. On the other hand the interface requires root and as such should allow what the admin asks for. Root inside a namespaces makes this more complicated because those might be not trusted in general. If they are not then such namespaces should be restricted anyway. Therefore drop the __GFP_NORETRY and replace it by __GFP_ACCOUNT to enfore memcg constrains on it.
--- net/netfilter/x_tables.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-)
net/netfilter/x_tables.c +1 -6
@@ -1178,12 +1178,7 @@ struct xt_table_info *xt_alloc_table_infif (sz <sizeof (*info) || sz >=XT_MAX_TABLE_SIZE )return NULL ; -/* __GFP_NORETRY is not fully supported by kvmalloc but it should -* work reasonably well if sz is too large and bail out rather -* than shoot all processes down before realizing there is nothing -* more to reclaim. -*/ - info =kvmalloc (sz,GFP_KERNEL | __GFP_NORETRY); + info =kvmalloc (sz,GFP_KERNEL_ACCOUNT );if (!info)return NULL ;
Greg Kroah-Hartman
maintainerPATCH 74/197 platform/x86: asus-nb-wmi: Add keymap entry for lid flip action on UX360
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aleh Filipovich <[email protected]>
[ Upstream commit 880b29ac107d15644bf4da228376ba3cd6af6d71 ]
Add entry to WMI keymap for lid flip event on Asus UX360.
On Asus Zenbook ux360 flipping lid from/to tablet mode triggers keyscan code 0xfa which cannot be handled and results in kernel log message "Unknown key fa pressed".
--- drivers/platform/x86/asus-nb-wmi.c | 1 + 1 file changed, 1 insertion(+)
drivers/platform/x86/asus-nb-wmi.c +1 -0
@@ -496,6 +496,7 @@ static const struct key_entry asus_nb_wm { KE_KEY, 0xC4, { KEY_KBDILLUMUP } }, { KE_KEY, 0xC5, { KEY_KBDILLUMDOWN } }, { KE_IGNORE, 0xC6, }, /* Ambient Light Sensor notification */ + { KE_KEY, 0xFA, { KEY_PROG2 } }, /* Lid flip action */ { KE_END, 0}, };
Greg Kroah-Hartman
maintainerPATCH 75/197 netfilter: fix memory leaks on netlink_dump_start error
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Westphal <[email protected]>
[ Upstream commit 3e673b23b541b8e7f773b2d378d6eb99831741cd ]
Shaochun Chen points out we leak dumper filter state allocations stored in dump_control->data in case there is an error before netlink sets cb_running (after which ->done will be called at some point).
In order to fix this, add .start functions and move allocations there.
Same pattern as used in commit 90fd131afc565159c9e0ea742f082b337e10f8c6 ("netfilter: nf_tables: move dumper state allocation into ->start").
--- net/netfilter/nf_conntrack_netlink.c | 26 +++++++++++++++++--------- net/netfilter/nfnetlink_acct.c | 29 +++++++++++++---------------- 2 files changed, 30 insertions(+), 25 deletions(-)
net/netfilter/nf_conntrack_netlink.c +17 -9
@@ -821,6 +821,21 @@ ctnetlink_alloc_filter(const struct nlat#endif } +static int ctnetlink_start (struct netlink_callback *cb) +{ +const struct nlattr *const *cda = cb->data; +struct ctnetlink_filter *filter =NULL ; + +if (cda[CTA_MARK ] && cda[CTA_MARK_MASK ]) { + filter =ctnetlink_alloc_filter (cda); +if (IS_ERR (filter)) +return PTR_ERR (filter); + } + + cb->data = filter; +return 0 ; +} +static int ctnetlink_filter_match (struct nf_conn *ct,void *data) {struct ctnetlink_filter *filter = data;
@@ -1240,19 +1255,12 @@ static int ctnetlink_get_conntrack(strucif (nlh->nlmsg_flags &NLM_F_DUMP ) {struct netlink_dump_control c = { + .start = ctnetlink_start, .dump = ctnetlink_dump_table, .done = ctnetlink_done, + .data = (void *)cda, }; -if (cda[CTA_MARK ] && cda[CTA_MARK_MASK ]) { -struct ctnetlink_filter *filter; - - filter =ctnetlink_alloc_filter (cda); -if (IS_ERR (filter)) -return PTR_ERR (filter); - - c.data = filter; - }return netlink_dump_start (ctnl, skb, nlh, &c); }
net/netfilter/nfnetlink_acct.c +13 -16
@@ -238,29 +238,33 @@ static const struct nla_policy filter_po [NFACCT_FILTER_VALUE] = { .type = NLA_U32 }, }; -static struct nfacct_filter * -nfacct_filter_alloc (const struct nlattr *const attr) +static int nfnl_acct_start (struct netlink_callback *cb) { -struct nfacct_filter *filter; +const struct nlattr *const attr = cb->data;struct nlattr *tb[NFACCT_FILTER_MAX +1 ]; +struct nfacct_filter *filter;int err; +if (!attr) +return 0 ; + err =nla_parse_nested (tb,NFACCT_FILTER_MAX , attr, filter_policy,NULL );if (err <0 ) -return ERR_PTR (err); +return err;if (!tb[NFACCT_FILTER_MASK ] || !tb[NFACCT_FILTER_VALUE ]) -return ERR_PTR (-EINVAL ); +return -EINVAL ; filter =kzalloc (sizeof (struct nfacct_filter ),GFP_KERNEL );if (!filter) -return ERR_PTR (-ENOMEM ); +return -ENOMEM ; filter->mask =ntohl (nla_get_be32 (tb[NFACCT_FILTER_MASK ])); filter->value =ntohl (nla_get_be32 (tb[NFACCT_FILTER_VALUE ])); + cb->data = filter; -return filter; +return 0 ; } static int nfnl_acct_get(struct net *net, struct sock *nfnl,
@@ -275,18 +279,11 @@ static int nfnl_acct_get(struct net *netif (nlh->nlmsg_flags &NLM_F_DUMP ) {struct netlink_dump_control c = { .dump = nfnl_acct_dump, + .start = nfnl_acct_start, .done = nfnl_acct_done, + .data = (void *)tb[NFACCT_FILTER ], }; -if (tb[NFACCT_FILTER ]) { -struct nfacct_filter *filter; - - filter =nfacct_filter_alloc (tb[NFACCT_FILTER ]); -if (IS_ERR (filter)) -return PTR_ERR (filter); - - c.data = filter; - }return netlink_dump_start (nfnl, skb, nlh, &c); }
Greg Kroah-Hartman
maintainerPATCH 76/197 tcp, ulp: add alias for all ulp modules
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Borkmann <[email protected]>
[ Upstream commit 037b0b86ecf5646f8eae777d8b52ff8b401692ec ]
Lets not turn the TCP ULP lookup into an arbitrary module loader as we only intend to load ULP modules through this mechanism, not other unrelated kernel modules:
[root@bar]# cat foo.c #include <sys/types.h> #include <sys/socket.h> #include <linux/tcp.h> #include <linux/in.h>
int main(void)
{
int sock = socket(PF_INET, SOCK_STREAM, 0);
setsockopt(sock, IPPROTO_TCP, TCP_ULP, "sctp", sizeof("sctp"));
return 0;
}[root@bar]# gcc foo.c -O2 -Wall [root@bar]# lsmod | grep sctp [root@bar]# ./a.out [root@bar]# lsmod | grep sctp sctp 1077248 4 libcrc32c 16384 3 nf_conntrack,nf_nat,sctp [root@bar]#
Fix it by adding module alias to TCP ULP modules, so probing module via request_module() will be limited to tcp-ulp-[name]. The existing modules like kTLS will load fine given tcp-ulp-tls alias, but others will fail to load:
[root@bar]# lsmod | grep sctp [root@bar]# ./a.out [root@bar]# lsmod | grep sctp [root@bar]#
Sockmap is not affected from this since it's either built-in or not.
--- include/net/tcp.h | 4 ++++ net/ipv4/tcp_ulp.c | 2 +- net/tls/tls_main.c | 1 + 3 files changed, 6 insertions(+), 1 deletion(-)
include/net/tcp.h +4 -0
@@ -2023,6 +2023,10 @@ int tcp_set_ulp_id(struct sock *sk, consvoid tcp_get_available_ulp (char *buf,size_t len);void tcp_cleanup_ulp (struct sock *sk); +#define MODULE_ALIAS_TCP_ULP (name) \ + __MODULE_INFO(alias, alias_userspace, name); \ + __MODULE_INFO(alias, alias_tcp_ulp, "tcp-ulp-" name) +/* Call BPF_SOCK_OPS program that returns an int. If the return value * is < 0, then the BPF op failed (for example if the loaded BPF * program does not support the chosen operation or there is no BPF
net/ipv4/tcp_ulp.c +1 -1
@@ -51,7 +51,7 @@ static const struct tcp_ulp_ops *__tcp_u#ifdef CONFIG_MODULES if (!ulp &&capable (CAP_NET_ADMIN )) {rcu_read_unlock (); -request_module ("%s" , name); +request_module ("tcp-ulp-%s" , name);rcu_read_lock (); ulp =tcp_ulp_find (name); }
net/tls/tls_main.c +1 -0
@@ -45,6 +45,7 @@MODULE_AUTHOR ("Mellanox Technologies" );MODULE_DESCRIPTION ("Transport Layer Security Support" );MODULE_LICENSE ("Dual BSD/GPL" ); +MODULE_ALIAS_TCP_ULP ("tls" );enum {TLSV4 ,
Greg Kroah-Hartman
maintainerPATCH 77/197 ubi: Initialize Fastmap checkmapping correctly
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Richard Weinberger <[email protected]>
[ Upstream commit 25677478474a91fa1b46f19a4a591a9848bca6fb ]
We cannot do it last, otherwithse it will be skipped for dynamic volumes.
--- drivers/mtd/ubi/vtbl.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-)
drivers/mtd/ubi/vtbl.c +10 -10
@@ -579,6 +579,16 @@ static int init_volumes(struct ubi_devic reserved_pebs += vol->reserved_pebs;/* +* We use ubi->peb_count and not vol->reserved_pebs because +* we want to keep the code simple. Otherwise we'd have to +* resize/check the bitmap upon volume resize too. +* Allocating a few bytes more does not hurt. +*/ + err =ubi_fastmap_init_checkmap (vol, ubi->peb_count); +if (err) +return err; + +/* * In case of dynamic volume UBI knows nothing about how many * data is stored there. So assume the whole volume is used. */
@@ -620,16 +630,6 @@ static int init_volumes(struct ubi_devic (long long )(vol->used_ebs -1 ) * vol->usable_leb_size; vol->used_bytes += av->last_data_size; vol->last_eb_bytes = av->last_data_size; - -/* -* We use ubi->peb_count and not vol->reserved_pebs because -* we want to keep the code simple. Otherwise we'd have to -* resize/check the bitmap upon volume resize too. -* Allocating a few bytes more does not hurt. -*/ - err =ubi_fastmap_init_checkmap (vol, ubi->peb_count); -if (err) -return err; }/* And add the layout volume */
Greg Kroah-Hartman
maintainerPATCH 78/197 RDMA/hns: Fix usage of bitmap allocation functions return values
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gal Pressman <[email protected]>
[ Upstream commit a1ceeca679dccc492235f0f629d9e9f7b3d51ca8 ]
hns bitmap allocation functions return 0 on success and -1 on failure. Callers of these functions wrongly used their return value as an errno, fix that by making a proper conversion.
--- drivers/infiniband/hw/hns/hns_roce_pd.c | 2 +- drivers/infiniband/hw/hns/hns_roce_qp.c | 5 ++++- 2 files changed, 5 insertions(+), 2 deletions(-)
drivers/infiniband/hw/hns/hns_roce_pd.c +1 -1
@@ -37,7 +37,7 @@static int hns_roce_pd_alloc (struct hns_roce_dev *hr_dev,unsigned long *pdn) { -return hns_roce_bitmap_alloc (&hr_dev->pd_bitmap, pdn); +return hns_roce_bitmap_alloc (&hr_dev->pd_bitmap, pdn) ? -ENOMEM :0 ; }static void hns_roce_pd_free (struct hns_roce_dev *hr_dev,unsigned long pdn)
drivers/infiniband/hw/hns/hns_roce_qp.c +4 -1
@@ -115,7 +115,10 @@ static int hns_roce_reserve_range_qp(str {struct hns_roce_qp_table *qp_table = &hr_dev->qp_table; -return hns_roce_bitmap_alloc_range (&qp_table->bitmap, cnt, align, base); +return hns_roce_bitmap_alloc_range (&qp_table->bitmap, cnt, align, + base) ? + -ENOMEM : +0 ; }enum hns_roce_qp_state to_hns_roce_state (enum ib_qp_state state)
Greg Kroah-Hartman
maintainerPATCH 79/197 ACPICA: ACPICA: add status check for acpi_hw_read before assigning return value
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Erik Schmauss <[email protected]>
[ Upstream commit f016b19a9275089a2ab06c2144567c2ad8d5d6ad ]
The value coming from acpi_hw_read() should not be used if it returns an error code, so check the status returned by it before using that value in two places in acpi_hw_register_read().
--- drivers/acpi/acpica/hwregs.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/acpi/acpica/hwregs.c +7 -2
@@ -528,13 +528,18 @@ acpi_status acpi_hw_register_read(u32 re status =acpi_hw_read (&value64, &acpi_gbl_FADT.xpm2_control_block); - value = (u32 )value64; +if (ACPI_SUCCESS (status)) { + value = (u32 )value64; + }break ;case ACPI_REGISTER_PM_TIMER :/* 32-bit access */ status =acpi_hw_read (&value64, &acpi_gbl_FADT.xpm_timer_block); - value = (u32 )value64; +if (ACPI_SUCCESS (status)) { + value = (u32 )value64; + } +break ;case ACPI_REGISTER_SMI_COMMAND_BLOCK :/* 8-bit access */
Greg Kroah-Hartman
maintainerPATCH 80/197 perf arm spe: Fix uninitialized record error variable
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kim Phillips <[email protected]>
[ Upstream commit 344353366591acf659a0d0dea498611da78d67e2 ]
The auxtrace init variable 'err' was not being initialized, leading perf to abort early in an SPE record command when there was no explicit error, rather only based whatever memory contents were on the stack. Initialize it explicitly on getting an SPE successfully, the same way cs-etm does.
--- tools/perf/arch/arm64/util/arm-spe.c | 1 + 1 file changed, 1 insertion(+)
tools/perf/arch/arm64/util/arm-spe.c +1 -0
@@ -194,6 +194,7 @@ struct auxtrace_record *arm_spe_recordin sper->itr.read_finish = arm_spe_read_finish; sper->itr.alignment =0 ; + *err =0 ;return &sper->itr; }
Greg Kroah-Hartman
maintainerPATCH 81/197 net: hns3: Fix for command format parsing error in hclge_is_all_function_id_zero
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xi Wang <[email protected]>
[ Upstream commit 6c39d5278e62956238a681e4cfc69fae5507fc57 ]
According to the functional specification of hardware, the first descriptor of response from command 'lookup vlan talbe' is not valid. Currently, the first descriptor is parsed as normal value, which will cause an expected error.
This patch fixes this problem by skipping the first descriptor.
--- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +1 -1
@@ -3911,7 +3911,7 @@ static bool hclge_is_all_function_id_zer#define HCLGE_FUNC_NUMBER_PER_DESC 6int i, j; -for (i =0 ; i <HCLGE_DESC_NUMBER ; i++) +for (i =1 ; i <HCLGE_DESC_NUMBER ; i++)for (j =0 ; j <HCLGE_FUNC_NUMBER_PER_DESC ; j++)if (desc[i].data[j])return false;
Greg Kroah-Hartman
maintainerPATCH 82/197 block: dont warn for flush on read-only device
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
[ Upstream commit b089cfd95d32638335c551651a8e00fd2c4edb0b ]
Don't warn for a flush issued to a read-only device. It's not strictly a writable command, as it doesn't change any on-media data by itself.
--- block/blk-core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
block/blk-core.c +3 -1
@@ -2159,7 +2159,9 @@ static inline bool should_fail_request(sstatic inline bool bio_check_ro (struct bio *bio,struct hd_struct *part) { -if (part->policy &&op_is_write (bio_op (bio))) { +const int op =bio_op (bio); + +if (part->policy && (op_is_write (op) && !op_is_flush (op))) {char b[BDEVNAME_SIZE ]; WARN_ONCE(1,
Re: [PATCH 4.18 082/197] block: dont warn for flush on read-only device
Hi Greg,
The follow up patch 8b2ded1c94c0 ("block: don't warn when doing fsync on read-only devices") should get applied too, since it correctly fixes what this patch tried to fix.
--
Stefan
Quoted 33 lines · Greg Kroah-Hartman
4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
[ Upstream commit
b089cfd95d32638335c551651a8e00fd2c4edb0b]Don't warn for a flush issued to a read-only device. It's not strictly a writable command, as it doesn't change any on-media data by itself.
Reported-by Stefan Agner <[email protected]>Fixes 721c7fc701c7 ("block: fail op_is_write() requests to read-only partitions")Signed-off-by Jens Axboe <[email protected]>Signed-off-by Sasha Levin <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>--- block/blk-core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)block/blk-core.c +3
-1@@ -2159,7 +2159,9 @@ static inline bool should_fail_request(sstatic inline bool bio_check_ro (struct bio *bio,struct hd_struct *part) { -if (part->policy &&op_is_write (bio_op (bio))) { +const int op =bio_op (bio); + +if (part->policy && (op_is_write (op) && !op_is_flush (op))) {char b[BDEVNAME_SIZE ]; WARN_ONCE(1,
Greg Kroah-Hartman
maintainerRe: [PATCH 4.18 082/197] block: dont warn for flush on read-only device
Quoted 5 lines · Stefan Agner
Hi Greg,
The follow up patch
8b2ded1c94c0("block: don't warn when doing fsync on read-only devices") should get applied too, since it correctly fixes what this patch tried to fix.
Already queueud up, thanks!
greg k-h
Greg Kroah-Hartman
maintainerPATCH 83/197 net: hns3: Fix for phy link issue when using marvell phy driver
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jian Shen <[email protected]>
[ Upstream commit 60081dcc4fce385ade26d3145b2479789df0b7e5 ]
For marvell phy m88e1510, bit SUPPORTED_FIBRE of phydev->supported is default on. Both phy_resume() and phy_suspend() will check the SUPPORTED_FIBRE bit and write register of fibre page.
Currently in hns3 driver, the SUPPORTED_FIBRE bit will be cleared after phy_connect_direct() finished. Because phy_resume() is called in phy_connect_direct(), and phy_suspend() is called when disconnect phy device, so the operation for fibre page register is not symmetrical. It will cause phy link issue when reload hns3 driver.
This patch fixes it by disable the SUPPORTED_FIBRE before connecting phy.
--- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c +2 -0
@@ -208,6 +208,8 @@ int hclge_mac_start_phy(struct hclge_devif (!phydev)return 0 ; + phydev->supported &= ~SUPPORTED_FIBRE ; + ret =phy_connect_direct (netdev, phydev, hclge_mac_adjust_link,PHY_INTERFACE_MODE_SGMII );
Greg Kroah-Hartman
maintainerPATCH 84/197 PCI: Match Root Ports MPS to endpoints MPSS as necessary
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Myron Stowe <[email protected]>
[ Upstream commit 9f0e89359775ee21fe1ea732e34edb52aef5addf ]
In commit 27d868b5e6cf ("PCI: Set MPS to match upstream bridge"), we made sure every device's MPS setting matches its upstream bridge, making it more likely that a hot-added device will work in a system with an optimized MPS configuration.
Recently I've started encountering systems where the endpoint device's MPSS capability is less than its Root Port's current MPS value, thus the endpoint is not capable of matching its upstream bridge's MPS setting (see:
bugzilla via "Link:" below). This leaves the system vulnerable - the upstream Root Port could respond with larger TLPs than the device can handle, and the device will consider them to be 'Malformed'.
One could use the "pci=pcie_bus_safe" kernel parameter to work around the issue, but that forces a user to supply a kernel parameter to get the system to function reliably and may end up limiting MPS settings of other unrelated, sub-topologies which could benefit from maintaining their larger values.
Augment Keith's approach to include tuning down a Root Port's MPS setting when its hot-added endpoint device is not capable of matching it.
--- drivers/pci/probe.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)
drivers/pci/probe.c +10 -2
@@ -1725,7 +1725,7 @@ int pci_setup_device(struct pci_dev *devstatic void pci_configure_mps (struct pci_dev *dev) {struct pci_dev *bridge =pci_upstream_bridge (dev); -int mps, p_mps, rc; +int mps, mpss, p_mps, rc;if (!pci_is_pcie (dev) || !bridge || !pci_is_pcie (bridge))return ;
@@ -1753,6 +1753,14 @@ static void pci_configure_mps(struct pciif (pcie_bus_config !=PCIE_BUS_DEFAULT )return ; + mpss =128 << dev->pcie_mpss; +if (mpss < p_mps &&pci_pcie_type (bridge) ==PCI_EXP_TYPE_ROOT_PORT ) { +pcie_set_mps (bridge, mpss); +pci_info (dev,"Upstream bridge's Max Payload Size set to %d (was %d, max %d)\n" , + mpss, p_mps,128 << bridge->pcie_mpss); + p_mps =pcie_get_mps (bridge); + } + rc =pcie_set_mps (dev, p_mps);if (rc) { pci_warn(dev, "can't set Max Payload Size to %d; if necessary, use \"pci=pcie_bus_safe\" and report a bug\n",
@@ -1761,7 +1769,7 @@ static void pci_configure_mps(struct pci }pci_info (dev,"Max Payload Size set to %d (was %d, max %d)\n" , - p_mps, mps,128 << dev->pcie_mpss); + p_mps, mps, mpss); }static struct hpp_type0 pci_default_type0 = {
Greg Kroah-Hartman
maintainerPATCH 85/197 drm/amd/display: Guard against null crtc in CRC IRQ
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Nicholas Kazlauskas <[email protected]>
[ Upstream commit dddc0557e3a02499ce336b1e2e67f5afaecccc80 ]
[Why]
A null pointer deference can occur if crtc is null in amdgpu_dm_crtc_handle_crc_irq. This can happen if get_crtc_by_otg_inst returns NULL during dm_crtc_high_irq, leading to a hang in some IGT test cases.
[How]
Check that CRTC is non-null before accessing its fields.
--- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c +8 -2
@@ -98,10 +98,16 @@ int amdgpu_dm_crtc_set_crc_source(struct*/ void amdgpu_dm_crtc_handle_crc_irq (struct drm_crtc *crtc) { -struct dm_crtc_state *crtc_state =to_dm_crtc_state (crtc->state); -struct dc_stream_state *stream_state = crtc_state->stream; +struct dm_crtc_state *crtc_state; +struct dc_stream_state *stream_state;uint32_t crcs[3 ]; +if (crtc ==NULL ) +return ; + + crtc_state =to_dm_crtc_state (crtc->state); + stream_state = crtc_state->stream; +/* Early return if CRC capture is not enabled. */ if (!crtc_state->crc_enabled)return ;
Greg Kroah-Hartman
maintainerPATCH 86/197 coccicheck: return proper error code on fail
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Denis Efremov <[email protected]>
[ Upstream commit 512ddf7d7db056edfed3159ea7cb4e4a5eefddd4 ]
If coccicheck fails, it should return an error code distinct from zero to signal about an internal problem. Current code instead of exiting with the tool's error code returns the error code of 'echo "coccicheck failed"' which is almost always equals to zero, thus failing the original intention of alerting about a problem. This patch fixes the code.
Found by Linux Driver Verification project (linuxtesting.org).
--- scripts/coccicheck | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
scripts/coccicheck +3 -2
@@ -128,9 +128,10 @@ run_cmd_parmap() { fi echo $@ >>$DEBUG_FILE $@ 2>>$DEBUG_FILE - if [[ $? -ne 0 ]]; then + err=$? + if [[ $err -ne 0 ]]; then echo "coccicheck failed" - exit $? + exit $err fi }
Greg Kroah-Hartman
maintainerPATCH 87/197 perf tools: Check for null when copying nsinfo.
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Benno Evers <[email protected]>
[ Upstream commit 3f4417d693b43fa240ac8bde4487f67745ca23d8 ]
The argument to nsinfo__copy() was assumed to be valid, but some code paths exist that will lead to NULL being passed.
In particular, running 'perf script -D' on a perf.data file containing an PERF_RECORD_MMAP event associating the '[vdso]' dso with pid 0 earlier in the event stream will lead to a segfault.
Since all calling code is already checking for a non-null return value, just return NULL for this case as well.
--- tools/perf/util/namespaces.c | 3 +++ 1 file changed, 3 insertions(+)
tools/perf/util/namespaces.c +3 -0
@@ -139,6 +139,9 @@ struct nsinfo *nsinfo__copy(struct nsinf {struct nsinfo *nnsi; +if (nsi ==NULL ) +return NULL ; + nnsi =calloc (1 ,sizeof (*nnsi));if (nnsi !=NULL ) { nnsi->pid = nsi->pid;
Greg Kroah-Hartman
maintainerPATCH 88/197 f2fs: avoid race between zero_range and background GC
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chao Yu <[email protected]>
[ Upstream commit c7079853c859c910b9d047a37891b4aafb8f8dd7 ]
Thread A Background GC
- f2fs_zero_range
- truncate_pagecache_range
- gc_data_segment
- get_read_data_page
- move_data_page
- set_page_dirty
- set_cold_data
- f2fs_do_zero_range
- dn->data_blkaddr = NEW_ADDR;
- f2fs_set_data_blkaddr
Actually, we don't need to set dirty & checked flag on the page, since all valid data in the page should be zeroed by zero_range().
Use i_gc_rwsem[WRITE] to avoid such race condition.
--- fs/f2fs/file.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
fs/f2fs/file.c +9 -2
@@ -1295,8 +1295,6 @@ static int f2fs_zero_range(struct inodeif (ret) goto out_sem; -truncate_pagecache_range (inode, offset, offset + len -1 ); - pg_start = ((unsigned long long ) offset) >>PAGE_SHIFT ; pg_end = ((unsigned long long ) offset + len) >>PAGE_SHIFT ;
@@ -1326,12 +1324,19 @@ static int f2fs_zero_range(struct inodeunsigned int end_offset;pgoff_t end; +down_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]); + +truncate_pagecache_range (inode, + (loff_t )index <<PAGE_SHIFT , + ((loff_t )pg_end <<PAGE_SHIFT ) -1 ); +f2fs_lock_op (sbi);set_new_dnode (&dn, inode,NULL ,NULL ,0 ); ret =f2fs_get_dnode_of_data (&dn, index,ALLOC_NODE );if (ret) {f2fs_unlock_op (sbi); +up_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]); goto out; }
@@ -1340,7 +1345,9 @@ static int f2fs_zero_range(struct inode ret =f2fs_do_zero_range (&dn, index, end);f2fs_put_dnode (&dn); +f2fs_unlock_op (sbi); +up_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]);f2fs_balance_fs (sbi, dn.node_changed);
Greg Kroah-Hartman
maintainerPATCH 89/197 f2fs: fix avoid race between truncate and background GC
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chao Yu <[email protected]>
[ Upstream commit a33c150237a20d97a174243bc658c86502f9d370 ]
Thread A Background GC
- f2fs_setattr isize to 0
- truncate_setsize
- gc_data_segment
- f2fs_get_read_data_page page #0
- set_page_dirty
- set_cold_data
- f2fs_truncate
- f2fs_setattr isize to 4k
- read 4k <--- hit data in cached page #0
Above race condition can cause read out invalid data in a truncated page, fix it by i_gc_rwsem[WRITE] lock.
--- fs/f2fs/data.c | 4 ++++ fs/f2fs/file.c | 37 +++++++++++++++++++++++-------------- 2 files changed, 27 insertions(+), 14 deletions(-)
fs/f2fs/data.c +4 -0
@@ -2149,8 +2149,12 @@ static void f2fs_write_failed(struct addif (to > i_size) {down_write (&F2FS_I (inode)->i_mmap_sem); +down_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]); +truncate_pagecache (inode, i_size);f2fs_truncate_blocks (inode, i_size, true); + +up_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]);up_write (&F2FS_I (inode)->i_mmap_sem); } }
fs/f2fs/file.c +23 -14
@@ -782,22 +782,26 @@ int f2fs_setattr(struct dentry *dentry, }if (attr->ia_valid &ATTR_SIZE ) { -if (attr->ia_size <=i_size_read (inode)) { -down_write (&F2FS_I (inode)->i_mmap_sem); -truncate_setsize (inode, attr->ia_size); +bool to_smaller = (attr->ia_size <=i_size_read (inode)); + +down_write (&F2FS_I (inode)->i_mmap_sem); +down_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]); + +truncate_setsize (inode, attr->ia_size); + +if (to_smaller) err =f2fs_truncate (inode); -up_write (&F2FS_I (inode)->i_mmap_sem); -if (err) -return err; - }else { -/* -* do not trim all blocks after i_size if target size is -* larger than i_size. -*/ -down_write (&F2FS_I (inode)->i_mmap_sem); -truncate_setsize (inode, attr->ia_size); -up_write (&F2FS_I (inode)->i_mmap_sem); +/* +* do not trim all blocks after i_size if target size is +* larger than i_size. +*/ +up_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]); +up_write (&F2FS_I (inode)->i_mmap_sem); + +if (err) +return err; +if (!to_smaller) {/* should convert inline inode here */ if (!f2fs_may_inline_data (inode)) { err =f2fs_convert_inline_inode (inode);
@@ -944,13 +948,18 @@ static int punch_hole(struct inode *inod blk_start = (loff_t )pg_start <<PAGE_SHIFT ; blk_end = (loff_t )pg_end <<PAGE_SHIFT ; +down_write (&F2FS_I (inode)->i_mmap_sem); +down_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]); +truncate_inode_pages_range (mapping, blk_start, blk_end -1 );f2fs_lock_op (sbi); ret =f2fs_truncate_hole (inode, pg_start, pg_end);f2fs_unlock_op (sbi); + +up_write (&F2FS_I (inode)->i_gc_rwsem[WRITE ]);up_write (&F2FS_I (inode)->i_mmap_sem); } }
Greg Kroah-Hartman
maintainerPATCH 90/197 RISC-V: Use KBUILD_CFLAGS instead of KCFLAGS when building the vDSO
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Palmer Dabbelt <[email protected]>
[ Upstream commit 4938c79bd0f5f3650c8c2cd4cdc972f0a6962ce4 ]
If you use a 64-bit compiler to build a 32-bit kernel then you'll get an error when building the vDSO due to a library mismatch. The happens because the relevant "-march" argument isn't supplied to the GCC run that generates one of the vDSO intermediate files.
I'm not actually sure what the right thing to do here is as I'm not particularly familiar with the kernel build system. I poked the documentation and it appears that KCFLAGS is the correct thing to do (it's suggested that should be used when building modules), but we set KBUILD_CFLAGS in arch/riscv/Makefile.
This does at least fix the build error.
--- arch/riscv/kernel/vdso/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/riscv/kernel/vdso/Makefile +2 -2
@@ -52,8 +52,8 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE # Add -lgcc so rv32 gets static muldi3 and lshrdi3 definitions. # Make sure only to export the intended __vdso_xxx symbol offsets. quiet_cmd_vdsold = VDSOLD $@ - cmd_vdsold = $(CC) $(KCFLAGS) $(call cc-option, -no-pie) -nostdlib $(SYSCFLAGS_$(@F)) \ - -Wl,-T,$(filter-out FORCE,$^) -o [email protected] -lgcc && \ + cmd_vdsold = $(CC) $(KBUILD_CFLAGS) $(call cc-option, -no-pie) -nostdlib -nostartfiles $(SYSCFLAGS_$(@F)) \ + -Wl,-T,$(filter-out FORCE,$^) -o [email protected] && \ $(CROSS_COMPILE)objcopy \ $(patsubst %, -G __vdso_%, $(vdso-syms)) [email protected] $@
Greg Kroah-Hartman
maintainerPATCH 91/197 irqchip/stm32: Fix init error handling
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dan Carpenter <[email protected]>
[ Upstream commit 4096165d55218a6f58b6c2ebc5d2428aa0aa70e4 ]
If there are any errors in stm32_exti_host_init() then it leads to a NULL dereference in the callers. The function should clean up after itself.
--- drivers/irqchip/irq-stm32-exti.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-)
drivers/irqchip/irq-stm32-exti.c +13 -12
@@ -602,17 +602,24 @@ stm32_exti_host_data *stm32_exti_host_insizeof (struct stm32_exti_chip_data ), GFP_KERNEL);if (!host_data->chips_data) -return NULL ; + goto free_host_data; host_data->base =of_iomap (node,0 );if (!host_data->base) {pr_err ("%pOF: Unable to map registers\n" , node); -return NULL ; + goto free_chips_data; } stm32_host_data = host_data;return host_data; + +free_chips_data: +kfree (host_data->chips_data); +free_host_data: +kfree (host_data); + +return NULL ; } static struct
@@ -664,10 +671,8 @@ static int __init stm32_exti_init(conststruct irq_domain *domain; host_data =stm32_exti_host_init (drv_data, node); -if (!host_data) { - ret = -ENOMEM ; - goto out_free_mem; - } +if (!host_data) +return -ENOMEM ; domain =irq_domain_add_linear (node, drv_data->bank_nr *IRQS_PER_BANK , &irq_exti_domain_ops,NULL );
@@ -724,7 +729,6 @@ out_free_domain:irq_domain_remove (domain); out_unmap:iounmap (host_data->base); -out_free_mem:kfree (host_data->chips_data);kfree (host_data);return ret;
@@ -751,10 +755,8 @@ __init stm32_exti_hierarchy_init(const s } host_data =stm32_exti_host_init (drv_data, node); -if (!host_data) { - ret = -ENOMEM ; - goto out_free_mem; - } +if (!host_data) +return -ENOMEM ;for (i =0 ; i < drv_data->bank_nr; i++)stm32_exti_chip_init (host_data, i, node);
@@ -776,7 +778,6 @@ __init stm32_exti_hierarchy_init(const s out_unmap:iounmap (host_data->base); -out_free_mem:kfree (host_data->chips_data);kfree (host_data);return ret;
Greg Kroah-Hartman
maintainerPATCH 92/197 irqchip/bcm7038-l1: Hide cpu offline callback when building for !SMP
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jonas Gorski <[email protected]>
[ Upstream commit 0702bc4d2fe793018ad9aa0eb14bff7f526c4095 ]
When compiling bmips with SMP disabled, the build fails with:
drivers/irqchip/irq-bcm7038-l1.o: In function `bcm7038_l1_cpu_offline':drivers/irqchip/irq-bcm7038-l1.c:242: undefined reference to `irq_set_affinity_locked' make[5]: *** [vmlinux] Error 1
Fix this by adding and setting bcm7038_l1_cpu_offline only when actually compiling for SMP. It wouldn't have been used anyway, as it requires CPU_HOTPLUG, which in turn requires SMP.
--- drivers/irqchip/irq-bcm7038-l1.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/irqchip/irq-bcm7038-l1.c +4 -0
@@ -217,6 +217,7 @@ static int bcm7038_l1_set_affinity(strucreturn 0 ; } +#ifdef CONFIG_SMP static void bcm7038_l1_cpu_offline (struct irq_data *d) {struct cpumask *mask =irq_data_get_affinity_mask (d);
@@ -241,6 +242,7 @@ static void bcm7038_l1_cpu_offline(struc }irq_set_affinity_locked (d, &new_affinity, false); } +#endif static int __init bcm7038_l1_init_one(struct device_node *dn, unsigned int idx,
@@ -293,7 +295,9 @@ static struct irq_chip bcm7038_l1_irq_ch .irq_mask = bcm7038_l1_mask, .irq_unmask = bcm7038_l1_unmask, .irq_set_affinity = bcm7038_l1_set_affinity, +#ifdef CONFIG_SMP .irq_cpu_offline = bcm7038_l1_cpu_offline, +#endif }; static int bcm7038_l1_map(struct irq_domain *d, unsigned int virq,
Greg Kroah-Hartman
maintainerPATCH 93/197 net/9p/trans_fd.c: fix race by holding the lock
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tomas Bortoli <[email protected]>
[ Upstream commit 9f476d7c540cb57556d3cc7e78704e6cd5100f5f ]
It may be possible to run p9_fd_cancel() with a deleted req->req_list and incur in a double del. To fix hold the client->lock while changing the status, so the other threads will be synchronized.
To: Eric Van Hensbergen <[email protected]>
To: Ron Minnich <[email protected]>
To: Latchesar Ionkov <[email protected]>
--- net/9p/trans_fd.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
net/9p/trans_fd.c +5 -5
@@ -199,15 +199,14 @@ static void p9_mux_poll_stop(struct p9_cstatic void p9_conn_cancel (struct p9_conn *m,int err) {struct p9_req_t *req, *rtmp; -unsigned long flags;LIST_HEAD (cancel_list);p9_debug (P9_DEBUG_ERROR ,"mux %p err %d\n" , m, err); -spin_lock_irqsave (&m->client->lock, flags); +spin_lock (&m->client->lock);if (m->err) { -spin_unlock_irqrestore (&m->client->lock, flags); +spin_unlock (&m->client->lock);return ; }
@@ -219,7 +218,6 @@ static void p9_conn_cancel(struct p9_conlist_for_each_entry_safe (req, rtmp, &m->unsent_req_list, req_list) {list_move (&req->req_list, &cancel_list); } -spin_unlock_irqrestore (&m->client->lock, flags);list_for_each_entry_safe (req, rtmp, &cancel_list, req_list) {p9_debug (P9_DEBUG_ERROR ,"call back req %p\n" , req);
@@ -228,6 +226,7 @@ static void p9_conn_cancel(struct p9_con req->t_err = err;p9_client_cb (m->client, req,REQ_STATUS_ERROR ); } +spin_unlock (&m->client->lock); } static __poll_t
@@ -375,8 +374,9 @@ static void p9_read_work(struct work_strif (m->req->status !=REQ_STATUS_ERROR ) status =REQ_STATUS_RCVD ;list_del (&m->req->req_list); -spin_unlock (&m->client->lock); +/* update req->status while holding client->lock */ p9_client_cb (m->client, m->req, status); +spin_unlock (&m->client->lock); m->rc.sdata =NULL ; m->rc.offset =0 ; m->rc.capacity =0 ;
Greg Kroah-Hartman
maintainerPATCH 94/197 net/9p: fix error path of p9_virtio_probe
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jean-Philippe Brucker <[email protected]>
[ Upstream commit 92aef4675d5b1b55404e1532379e343bed0e5cf2 ]
Currently when virtio_find_single_vq fails, we go through del_vqs which throws a warning (Trying to free already-free IRQ). Skip del_vqs if vq allocation failed.
--- net/9p/trans_virtio.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/9p/trans_virtio.c +2 -1
@@ -571,7 +571,7 @@ static int p9_virtio_probe(struct virtio chan->vq =virtio_find_single_vq (vdev, req_done,"requests" );if (IS_ERR (chan->vq)) { err =PTR_ERR (chan->vq); - goto out_free_vq; + goto out_free_chan; } chan->vq->vdev->priv = chan;spin_lock_init (&chan->lock);
@@ -624,6 +624,7 @@ out_free_tag:kfree (tag); out_free_vq: vdev->config->del_vqs (vdev); +out_free_chan:kfree (chan); fail:return err;
Greg Kroah-Hartman
maintainerPATCH 95/197 f2fs: fix to clear PG_checked flag in set_page_dirty()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chao Yu <[email protected]>
[ Upstream commit 66110abc4c931f879d70e83e1281f891699364bf ]
PG_checked flag will be set on data page during GC, later, we can recognize such page by the flag and migrate page to cold segment.
But previously, we don't clear this flag when invalidating data page, after page redirtying, we will write it into wrong log.
Let's clear PG_checked flag in set_page_dirty() to avoid this.
--- fs/f2fs/data.c | 4 ++++ 1 file changed, 4 insertions(+)
fs/f2fs/data.c +4 -0
@@ -2494,6 +2494,10 @@ static int f2fs_set_data_page_dirty(struif (!PageUptodate (page))SetPageUptodate (page); +/* don't remain PG_checked flag which was set during GC */ +if (is_cold_data (page)) +clear_cold_data (page); +if (f2fs_is_atomic_file (inode) && !f2fs_is_commit_atomic_write (inode)) {if (!IS_ATOMIC_WRITTEN_PAGE (page)) {f2fs_register_inmem_page (inode, page);
Greg Kroah-Hartman
maintainerPATCH 96/197 pinctrl: axp209: Fix NULL pointer dereference after allocation
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anton Vasilyev <[email protected]>
[ Upstream commit 504c76979bccec66e4c2e41f6a006e49e284466f ]
There is no check that allocation in axp20x_funcs_groups_from_mask is successful.
The patch adds corresponding check and return values.
Found by Linux Driver Verification project (linuxtesting.org).
--- drivers/pinctrl/pinctrl-axp209.c | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-)
drivers/pinctrl/pinctrl-axp209.c +20 -6
@@ -316,7 +316,7 @@ static const struct pinctrl_ops axp20x_p .get_group_pins = axp20x_group_pins, }; -static void axp20x_funcs_groups_from_mask (struct device *dev,unsigned int mask, +static int axp20x_funcs_groups_from_mask (struct device *dev,unsigned int mask,unsigned int mask_len,struct axp20x_pinctrl_function *func,const struct pinctrl_pin_desc *pins)
@@ -331,18 +331,22 @@ static void axp20x_funcs_groups_from_mas func->groups =devm_kcalloc (dev, ngroups,sizeof (const char *),GFP_KERNEL ); +if (!func->groups) +return -ENOMEM ; group = func->groups;for_each_set_bit (bit, &mask_cpy, mask_len) { *group = pins[bit].name; group++; } } + +return 0 ; } -static void axp20x_build_funcs_groups (struct platform_device *pdev) +static int axp20x_build_funcs_groups (struct platform_device *pdev) {struct axp20x_pctl *pctl =platform_get_drvdata (pdev); -int i, pin, npins = pctl->desc->npins; +int i, ret, pin, npins = pctl->desc->npins; pctl->funcs[AXP20X_FUNC_GPIO_OUT ].name ="gpio_out" ; pctl->funcs[AXP20X_FUNC_GPIO_OUT ].muxval =AXP20X_MUX_GPIO_OUT ;
@@ -366,13 +370,19 @@ static void axp20x_build_funcs_groups(st pctl->funcs[i].groups[pin] = pctl->desc->pins[pin].name; } -axp20x_funcs_groups_from_mask (&pdev->dev, pctl->desc->ldo_mask, + ret =axp20x_funcs_groups_from_mask (&pdev->dev, pctl->desc->ldo_mask, npins, &pctl->funcs[AXP20X_FUNC_LDO ], pctl->desc->pins); +if (ret) +return ret; -axp20x_funcs_groups_from_mask (&pdev->dev, pctl->desc->adc_mask, + ret =axp20x_funcs_groups_from_mask (&pdev->dev, pctl->desc->adc_mask, npins, &pctl->funcs[AXP20X_FUNC_ADC ], pctl->desc->pins); +if (ret) +return ret; + +return 0 ; }static const struct of_device_id axp20x_pctl_match[] = {
@@ -424,7 +434,11 @@ static int axp20x_pctl_probe(struct platplatform_set_drvdata (pdev, pctl); -axp20x_build_funcs_groups (pdev); + ret =axp20x_build_funcs_groups (pdev); +if (ret) { +dev_err (&pdev->dev,"failed to build groups\n" ); +return ret; + } pctrl_desc =devm_kzalloc (&pdev->dev,sizeof (*pctrl_desc),GFP_KERNEL );if (!pctrl_desc)
Greg Kroah-Hartman
maintainerPATCH 97/197 bpf: fix bpffs non-array map seq_show issue
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yonghong Song <[email protected]>
[ Upstream commit dc1508a579e682a1e5f1ed0753390e0aa7c23a97 ]
In function map_seq_next() of kernel/bpf/inode.c, the first key will be the "0" regardless of the map type. This works for array. But for hash type, if it happens key "0" is in the map, the bpffs map show will miss some items if the key "0" is not the first element of the first bucket.
This patch fixed the issue by guaranteeing to get the first element, if the seq_show is just started, by passing NULL pointer key to map_get_next_key() callback. This way, no missing elements will occur for
bpffs hash table show even if key "0" is in the map.
--- kernel/bpf/inode.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
kernel/bpf/inode.c +5 -3
@@ -196,19 +196,21 @@ static void *map_seq_next(struct seq_fil {struct bpf_map *map =seq_file_to_map (m);void *key =map_iter (m)->key; +void *prev_key;if (map_iter (m)->done)return NULL ;if (unlikely (v ==SEQ_START_TOKEN )) - goto done; + prev_key =NULL ; +else + prev_key = key; -if (map->ops->map_get_next_key (map, key, key)) { +if (map->ops->map_get_next_key (map, prev_key, key)) {map_iter (m)->done = true;return NULL ; } -done: ++(*pos);return key; }
Greg Kroah-Hartman
maintainerPATCH 98/197 powerpc/uaccess: Enable get_user(u64, *p) on 32-bit
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michael Ellerman <[email protected]>
[ Upstream commit f7a6947cd49b7ff4e03f1b4f7e7b223003d752ca ]
Currently if you build a 32-bit powerpc kernel and use get_user() to load a u64 value it will fail to build with eg:
kernel/rseq.o: In function `rseq_get_rseq_cs':
kernel/rseq.c:123: undefined reference to `__get_user_bad'
This is hitting the check in __get_user_size() that makes sure the size we're copying doesn't exceed the size of the destination:
#define __get_user_size(x, ptr, size, retval)
do {
retval = 0;
__chk_user_ptr(ptr);
if (size > sizeof(x))
(x) = __get_user_bad();Which doesn't immediately make sense because the size of the destination is u64, but it's not really, because __get_user_check() etc. internally create an unsigned long and copy into that:
#define __get_user_check(x, ptr, size)
({
long __gu_err = -EFAULT;
unsigned long __gu_val = 0;The problem being that on 32-bit unsigned long is not big enough to hold a u64. We can fix this with a trick from hpa in the x86 code, we statically check the type of x and set the type of __gu_val to either unsigned long or unsigned long long.
--- arch/powerpc/include/asm/uaccess.h | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-)
arch/powerpc/include/asm/uaccess.h +10 -3
@@ -250,10 +250,17 @@ do { \ } \ }while (0 ) +/* +* This is a type: either unsigned long, if the argument fits into +* that type, or otherwise unsigned long long. + */ +#define __long_type (x) \ + __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL)) +#define __get_user_nocheck (x, ptr, size) \ ({ \ long __gu_err; \ - unsigned long __gu_val; \ + __long_type(*(ptr)) __gu_val; \ const __typeof__(*(ptr)) __user *__gu_addr = (ptr); \ __chk_user_ptr(ptr); \ if (!is_kernel_addr((unsigned long)__gu_addr)) \
@@ -267,7 +274,7 @@ do { \#define __get_user_check (x, ptr, size) \ ({ \ long __gu_err = -EFAULT; \ - unsigned long __gu_val = 0; \ + __long_type(*(ptr)) __gu_val = 0; \ const __typeof__(*(ptr)) __user *__gu_addr = (ptr); \ might_fault(); \ if (access_ok(VERIFY_READ, __gu_addr, (size))) { \
@@ -281,7 +288,7 @@ do { \#define __get_user_nosleep (x, ptr, size) \ ({ \ long __gu_err; \ - unsigned long __gu_val; \ + __long_type(*(ptr)) __gu_val; \ const __typeof__(*(ptr)) __user *__gu_addr = (ptr); \ __chk_user_ptr(ptr); \ barrier_nospec(); \
Greg Kroah-Hartman
maintainerPATCH 99/197 powerpc: Fix size calculation using resource_size()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dan Carpenter <[email protected]>
[ Upstream commit c42d3be0c06f0c1c416054022aa535c08a1f9b39 ]
The problem is the the calculation should be "end - start + 1" but the plus one is missing in this calculation.
--- arch/powerpc/sysdev/mpic_msgr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/sysdev/mpic_msgr.c +1 -1
@@ -196,7 +196,7 @@ static int mpic_msgr_probe(struct platfo/* IO map the message register block. */ of_address_to_resource (np,0 , &rsrc); - msgr_block_addr =ioremap (rsrc.start, rsrc.end - rsrc.start); + msgr_block_addr =ioremap (rsrc.start,resource_size (&rsrc));if (!msgr_block_addr) {dev_err (&dev->dev,"Failed to iomap MPIC message registers" );return -EFAULT ;
Greg Kroah-Hartman
maintainerPATCH 100/197 perf probe powerpc: Fix trace event post-processing
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sandipan Das <[email protected]>
[ Upstream commit 354b064b8ebc1e1ede58550ca9e08bfa81e6af43 ]
In some cases, a symbol may have multiple aliases. Attempting to add an entry probe for such symbols results in a probe being added at an incorrect location while it fails altogether for return probes. This is only applicable for binaries with debug information.
During the arch-dependent post-processing, the offset from the start of the symbol at which the probe is to be attached is determined and added to the start address of the symbol to get the probe's location. In case there are multiple aliases, this offset gets added multiple times for each alias of the symbol and we end up with an incorrect probe location.
This can be verified on a powerpc64le system as shown below.
$ nm /lib/modules/$(uname -r)/build/vmlinux | grep "sys_open$" ... c000000000414290 T __se_sys_open c000000000414290 T sys_open
$ objdump -d /lib/modules/$(uname -r)/build/vmlinux | grep -A 10 "<__se_sys_open>:"
c000000000414290 <__se_sys_open>:
c000000000414290: 19 01 4c 3c addis r2,r12,281 c000000000414294: 70 c4 42 38 addi r2,r2,-15248 c000000000414298: a6 02 08 7c mflr r0 c00000000041429c: e8 ff a1 fb std r29,-24(r1) c0000000004142a0: f0 ff c1 fb std r30,-16(r1) c0000000004142a4: f8 ff e1 fb std r31,-8(r1) c0000000004142a8: 10 00 01 f8 std r0,16(r1) c0000000004142ac: c1 ff 21 f8 stdu r1,-64(r1) c0000000004142b0: 78 23 9f 7c mr r31,r4 c0000000004142b4: 78 1b 7e 7c mr r30,r3
For both the entry probe and the return probe, the probe location should be _text+4276888 (0xc000000000414298). Since another alias exists for 'sys_open', the post-processing code will end up adding the offset (8 for powerpc64le) twice and perf will attempt to add the probe at _text+4276896 (0xc0000000004142a0) instead.
Before:
# perf probe -v -a sys_open
probe-definition(0): sys_open
symbol:sys_open file:(null) line:0 offset:0 return:0 lazy:(null) 0 arguments
Looking at the vmlinux_path (8 entries long)
Using /lib/modules/4.18.0-rc8+/build/vmlinux for symbols Open Debuginfo file: /lib/modules/4.18.0-rc8+/build/vmlinux Try to find probe point from debuginfo.
Symbol sys_open address found : c000000000414290 Matched function: __se_sys_open [2ad03a0]
Probe point found: __se_sys_open+0
Found 1 probe_trace_events.
Opening /sys/kernel/debug/tracing/kprobe_events write=1 Writing event: p:probe/sys_open _text+4276896
Added new event:
probe:sys_open (on sys_open)
...
# perf probe -v -a sys_open%return $retval
probe-definition(0): sys_open%return
symbol:sys_open file:(null) line:0 offset:0 return:1 lazy:(null) 0 arguments
Looking at the vmlinux_path (8 entries long)
Using /lib/modules/4.18.0-rc8+/build/vmlinux for symbols Open Debuginfo file: /lib/modules/4.18.0-rc8+/build/vmlinux Try to find probe point from debuginfo.
Symbol sys_open address found : c000000000414290
Matched function: __se_sys_open [2ad03a0]
Probe point found: __se_sys_open+0
Found 1 probe_trace_events.
Opening /sys/kernel/debug/tracing/README write=0
Opening /sys/kernel/debug/tracing/kprobe_events write=1 Parsing probe_events: p:probe/sys_open _text+4276896 Group:probe Event:sys_open probe:p
Writing event: r:probe/sys_open__return _text+4276896 Failed to write event: Invalid argument
Error: Failed to add events. Reason: Invalid argument (Code: -22)
After:
# perf probe -v -a sys_open
probe-definition(0): sys_open
symbol:sys_open file:(null) line:0 offset:0 return:0 lazy:(null) 0 arguments
Looking at the vmlinux_path (8 entries long)
Using /lib/modules/4.18.0-rc8+/build/vmlinux for symbols Open Debuginfo file: /lib/modules/4.18.0-rc8+/build/vmlinux Try to find probe point from debuginfo.
Symbol sys_open address found : c000000000414290 Matched function: __se_sys_open [2ad03a0]
Probe point found: __se_sys_open+0
Found 1 probe_trace_events.
Opening /sys/kernel/debug/tracing/kprobe_events write=1 Writing event: p:probe/sys_open _text+4276888
Added new event:
probe:sys_open (on sys_open)
...
# perf probe -v -a sys_open%return $retval
probe-definition(0): sys_open%return
symbol:sys_open file:(null) line:0 offset:0 return:1 lazy:(null) 0 arguments
Looking at the vmlinux_path (8 entries long)
Using /lib/modules/4.18.0-rc8+/build/vmlinux for symbols Open Debuginfo file: /lib/modules/4.18.0-rc8+/build/vmlinux Try to find probe point from debuginfo.
Symbol sys_open address found : c000000000414290 Matched function: __se_sys_open [2ad03a0]
Probe point found: __se_sys_open+0
Found 1 probe_trace_events.
Opening /sys/kernel/debug/tracing/README write=0 Opening /sys/kernel/debug/tracing/kprobe_events write=1 Parsing probe_events: p:probe/sys_open _text+4276888 Group:probe Event:sys_open probe:p
Writing event: r:probe/sys_open__return _text+4276888 Added new event:
probe:sys_open__return (on sys_open%return)
...
--- tools/perf/arch/powerpc/util/sym-handling.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
tools/perf/arch/powerpc/util/sym-handling.c +3 -1
@@ -141,8 +141,10 @@ void arch__post_process_probe_trace_evenfor (i =0 ; i < ntevs; i++) { tev = &pev->tevs[i];map__for_each_symbol (map, sym, tmp) { -if (map->unmap_ip (map, sym->start) == tev->point.address) +if (map->unmap_ip (map, sym->start) == tev->point.address) {arch__fix_tev_from_maps (pev, tev, map, sym); +break ; + } } } }
Greg Kroah-Hartman
maintainerPATCH 101/197 block: bvec_nr_vecs() returns value for wrong slab
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Greg Edwards <[email protected]>
[ Upstream commit d6c02a9beb67f13d5f14f23e72fa9981e8b84477 ]
In commit ed996a52c868 ("block: simplify and cleanup bvec pool handling"), the value of the slab index is incremented by one in bvec_alloc() after the allocation is done to indicate an index value of 0 does not need to be later freed.
bvec_nr_vecs() was not updated accordingly, and thus returns the wrong value. Decrement idx before performing the lookup.
--- block/bio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
block/bio.c +1 -1
@@ -156,7 +156,7 @@ out:unsigned int bvec_nr_vecs (unsigned short idx) { -return bvec_slabs[idx].nr_vecs; +return bvec_slabs[--idx].nr_vecs; }void bvec_free (mempool_t *pool,struct bio_vec *bv,unsigned int idx)
Greg Kroah-Hartman
maintainerPATCH 102/197 brcmfmac: fix brcmf_wiphy_wowl_params() NULL pointer dereference
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Winnie Chang <[email protected]>
[ Upstream commit 27a8aea13053700ad2a08189024df7e341d1ee51 ]
The kernel BUG happens when wowl is enabled from firmware. In brcmf_wiphy_wowl_params(), cfg is a NULL pointer because it is drvr->config returned from wiphy_to_cfg(), and drvr->config is not set yet. To fix it, set drvr->config before brcmf_setup_wiphy() which calls brcmf_wiphy_wowl_params().
--- drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +4 -4
@@ -6926,15 +6926,15 @@ struct brcmf_cfg80211_info *brcmf_cfg802 cfg->d11inf.io_type = (u8 )io_type;brcmu_d11_attach (&cfg->d11inf); - err =brcmf_setup_wiphy (wiphy, ifp); -if (err <0 ) - goto priv_out; -/* regulatory notifer below needs access to cfg so * assign it now. */ drvr->config = cfg; + err =brcmf_setup_wiphy (wiphy, ifp); +if (err <0 ) + goto priv_out; +brcmf_dbg (INFO ,"Registering custom regulatory\n" ); wiphy->reg_notifier = brcmf_cfg80211_reg_notifier; wiphy->regulatory_flags |=REGULATORY_CUSTOM_REG ;
Greg Kroah-Hartman
maintainerPATCH 103/197 s390/dasd: fix hanging offline processing due to canceled worker
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stefan Haberland <[email protected]>
[ Upstream commit 669f3765b755fd8739ab46ce3a9c6292ce8b3d2a ]
During offline processing two worker threads are canceled without freeing the device reference which leads to a hanging offline process.
--- drivers/s390/block/dasd_eckd.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/s390/block/dasd_eckd.c +5 -2
@@ -2035,8 +2035,11 @@ static int dasd_eckd_basic_to_ready(strustatic int dasd_eckd_online_to_ready (struct dasd_device *device) { -cancel_work_sync (&device->reload_device); -cancel_work_sync (&device->kick_validate); +if (cancel_work_sync (&device->reload_device)) +dasd_put_device (device); +if (cancel_work_sync (&device->kick_validate)) +dasd_put_device (device); +return 0 ; };
Greg Kroah-Hartman
maintainerPATCH 104/197 s390/dasd: fix panic for failed online processing
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stefan Haberland <[email protected]>
[ Upstream commit 7c6553d4db03350dad0110c3224194c19df76a8f ]
Fix a panic that occurs for a device that got an error in dasd_eckd_check_characteristics() during online processing. For example the read configuration data command may have failed.
If this error occurs the device is not being set online and the earlier invoked steps during online processing are rolled back. Therefore dasd_eckd_uncheck_device() is called which needs a valid private structure. But this pointer is not valid if dasd_eckd_check_characteristics() has failed.
Check for a valid device->private pointer to prevent a panic.
--- drivers/s390/block/dasd_eckd.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/s390/block/dasd_eckd.c +3 -0
@@ -1780,6 +1780,9 @@ static void dasd_eckd_uncheck_device(strstruct dasd_eckd_private *private = device->private;int i; +if (!private) +return ; +dasd_alias_disconnect_device_from_lcu (device); private->ned =NULL ; private->sneq =NULL ;
Greg Kroah-Hartman
maintainerPATCH 105/197 ACPI / scan: Initialize status to ACPI_STA_DEFAULT
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hans de Goede <[email protected]>
[ Upstream commit 5971b0c1594d6c34e257101ed5fdffec65205c50 ]
Since commit 63347db0affa "ACPI / scan: Use acpi_bus_get_status() to initialize ACPI_TYPE_DEVICE devs" the status field of normal acpi_devices gets set to 0 by acpi_bus_type_and_status() and filled with its actual value later when acpi_add_single_object() calls acpi_bus_get_status().
This means that any acpi_match_device_ids() calls in between will always fail with -ENOENT.
We already have a workaround for this, which temporary forces status to ACPI_STA_DEFAULT in drivers/acpi/x86/utils.c: acpi_device_always_present() and the next commit in this series adds another acpi_match_device_ids() call between status being initialized as 0 and the acpi_bus_get_status() call.
Rather then adding another workaround, this commit makes acpi_bus_type_and_status() initialize status to ACPI_STA_DEFAULT, this is safe to do as the only code looking at status between the initialization and the acpi_bus_get_status() call is those acpi_match_device_ids() calls.
Note this does mean that we need to (re)set status to 0 in case the acpi_bus_get_status() call fails.
--- drivers/acpi/scan.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/acpi/scan.c +3 -2
@@ -1612,7 +1612,8 @@ static int acpi_add_single_object(struct* Note this must be done before the get power-/wakeup_dev-flags calls. */ if (type ==ACPI_BUS_TYPE_DEVICE ) -acpi_bus_get_status (device); +if (acpi_bus_get_status (device) <0 ) +acpi_set_device_status (device,0 );acpi_bus_get_power_flags (device);acpi_bus_get_wakeup_device_flags (device);
@@ -1690,7 +1691,7 @@ static int acpi_bus_type_and_status(acpi* acpi_add_single_object updates this once we've an acpi_device * so that acpi_bus_get_status' quirk handling can be used. */ - *sta =0 ; + *sta =ACPI_STA_DEFAULT ;break ;case ACPI_TYPE_PROCESSOR : *type =ACPI_BUS_TYPE_PROCESSOR ;
Greg Kroah-Hartman
maintainerPATCH 106/197 blk-mq: count the hctx as active before allocating tag
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jianchao Wang <[email protected]>
[ Upstream commit d263ed9926823c462f99a7679e18f0c9e5b8550d ]
Currently, we count the hctx as active after allocate driver tag successfully. If a previously inactive hctx try to get tag first time, it may fails and need to wait. However, due to the stale tag ->active_queues, the other shared-tags users are still able to occupy all driver tags while there is someone waiting for tag. Consequently, even if the previously inactive hctx is waked up, it still may not be able to get a tag and could be starved.
To fix it, we count the hctx as active before try to allocate driver tag, then when it is waiting the tag, the other shared-tag users will reserve budget for it.
--- block/blk-mq-tag.c | 3 +++ block/blk-mq.c | 8 ++++++-- 2 files changed, 9 insertions(+), 2 deletions(-)
block/blk-mq-tag.c +3 -0
@@ -23,6 +23,9 @@ bool blk_mq_has_free_tags(struct blk_mq_/* * If a previously inactive queue goes active, bump the active user count. +* We need to do this before try to allocate driver tag, then even if fail +* to get tag when first time, the other shared-tag users could reserve +* budget for it. */ bool __blk_mq_tag_busy (struct blk_mq_hw_ctx *hctx) {
block/blk-mq.c +6 -2
@@ -285,7 +285,7 @@ static struct request *blk_mq_rq_ctx_ini rq->tag =-1 ; rq->internal_tag = tag; }else { -if (blk_mq_tag_busy (data->hctx)) { +if (data->hctx->flags &BLK_MQ_F_TAG_SHARED ) { rq_flags =RQF_MQ_INFLIGHT ;atomic_inc (&data->hctx->nr_active); }
@@ -367,6 +367,8 @@ static struct request *blk_mq_get_requesif (!op_is_flush (op) && e->type->ops.mq.limit_depth && !(data->flags &BLK_MQ_REQ_RESERVED )) e->type->ops.mq.limit_depth (op, data); + }else { +blk_mq_tag_busy (data->hctx); } tag =blk_mq_get_tag (data);
@@ -970,6 +972,7 @@ bool blk_mq_get_driver_tag(struct reques .hctx = blk_mq_map_queue(rq->q, rq->mq_ctx->cpu), .flags = wait ? 0 : BLK_MQ_REQ_NOWAIT, }; +bool shared;might_sleep_if (wait);
@@ -979,9 +982,10 @@ bool blk_mq_get_driver_tag(struct requesif (blk_mq_tag_is_reserved (data.hctx->sched_tags, rq->internal_tag)) data.flags |=BLK_MQ_REQ_RESERVED ; + shared =blk_mq_tag_busy (data.hctx); rq->tag =blk_mq_get_tag (&data);if (rq->tag >=0 ) { -if (blk_mq_tag_busy (data.hctx)) { +if (shared) { rq->rq_flags |=RQF_MQ_INFLIGHT ;atomic_inc (&data.hctx->nr_active); }
Greg Kroah-Hartman
maintainerPATCH 107/197 scsi: aic94xx: fix an error code in aic94xx_init()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dan Carpenter <[email protected]>
[ Upstream commit 0756c57bce3d26da2592d834d8910b6887021701 ]
We accidentally return success instead of -ENOMEM on this error path.
--- drivers/scsi/aic94xx/aic94xx_init.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/scsi/aic94xx/aic94xx_init.c +3 -1
@@ -1030,8 +1030,10 @@ static int __init aic94xx_init(void) aic94xx_transport_template =sas_domain_attach_transport (&aic94xx_transport_functions); -if (!aic94xx_transport_template) +if (!aic94xx_transport_template) { + err = -ENOMEM ; goto out_destroy_caches; + } err =pci_register_driver (&aic94xx_pci_driver);if (err)
Greg Kroah-Hartman
maintainerPATCH 108/197 NFSv4: Fix error handling in nfs4_sp4_select_mode()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Wei Yongjun <[email protected]>
[ Upstream commit 72bf75cfc00c02aa66ef6133048f37aa5d88825c ]
Error code is set in the error handling cases but never used. Fix it.
--- fs/nfs/nfs4proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/nfs/nfs4proc.c +1 -1
@@ -7734,7 +7734,7 @@ static int nfs4_sp4_select_mode(struct n } out: clp->cl_sp4_flags = flags; -return 0 ; +return ret; }struct nfs41_exchange_id_data {
Greg Kroah-Hartman
maintainerPATCH 109/197 Input: do not use WARN() in input_alloc_absinfo()
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dmitry Torokhov <[email protected]>
[ Upstream commit 100294cee9a98bfd4d6cb2d1c8a8aef0e959b0c4 ]
Some of fuzzers set panic_on_warn=1 so that they can handle WARN()ings the same way they handle full-blown kernel crashes. We used WARN() in input_alloc_absinfo() to get a better idea where memory allocation failed, but since then kmalloc() and friends started dumping call stack on memory allocation failures anyway, so we are not getting anything extra from WARN().
Because of the above, let's replace WARN with dev_err(). We use dev_err() instead of simply removing message and relying on kcalloc() to give us stack dump so that we'd know the instance of hardware device to which we were trying to attach input device.
--- drivers/input/input.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)
drivers/input/input.c +12 -4
@@ -480,11 +480,19 @@ EXPORT_SYMBOL(input_inject_event);*/ void input_alloc_absinfo (struct input_dev *dev) { -if (!dev->absinfo) - dev->absinfo =kcalloc (ABS_CNT ,sizeof (*dev->absinfo), -GFP_KERNEL ); +if (dev->absinfo) +return ; -WARN (!dev->absinfo,"%s(): kcalloc() failed?\n" , __func__); + dev->absinfo =kcalloc (ABS_CNT ,sizeof (*dev->absinfo),GFP_KERNEL ); +if (!dev->absinfo) { +dev_err (dev->dev.parent ?: &dev->dev, +"%s: unable to allocate memory\n" , __func__); +/* +* We will handle this allocation failure in +* input_register_device() when we refuse to register input +* device with ABS bits but without absinfo. +*/ + } }EXPORT_SYMBOL (input_alloc_absinfo);
Greg Kroah-Hartman
maintainerPATCH 110/197 xen/balloon: fix balloon initialization for PVH Dom0
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Roger Pau Monne <[email protected]>
[ Upstream commit 3596924a233e45aa918c961a902170fc4916461b ]
The current balloon code tries to calculate a delta factor for the balloon target when running in HVM mode in order to account for memory used by the firmware.
This workaround for memory accounting doesn't work properly on a PVH Dom0, that has a static-max value different from the target value even at startup. Note that this is not a problem for DomUs because guests are started with a static-max value that matches the amount of RAM in the memory map.
Fix this by forcefully setting target_diff for Dom0, regardless of it's mode.
--- drivers/xen/xen-balloon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/xen/xen-balloon.c +1 -1
@@ -81,7 +81,7 @@ static void watch_target(struct xenbus_w static_max = new_target; else static_max >>= PAGE_SHIFT - 10; - target_diff =xen_pv_domain () ?0 + target_diff = (xen_pv_domain () ||xen_initial_domain ()) ?0 : static_max - balloon_stats.target_pages; }
Greg Kroah-Hartman
maintainerPATCH 111/197 PCI: mvebu: Fix I/O space end address calculation
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Petazzoni <[email protected]>
[ Upstream commit dfd0309fd7b30a5baffaf47b2fccb88b46d64d69 ]
pcie->realio.end should be the address of last byte of the area, therefore using resource_size() of another resource is not correct, we must substract 1 to get the address of the last byte.
--- drivers/pci/controller/pci-mvebu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/pci/controller/pci-mvebu.c +1 -1
@@ -1219,7 +1219,7 @@ static int mvebu_pcie_probe(struct platf pcie->realio.start =PCIBIOS_MIN_IO ; pcie->realio.end =min_t (resource_size_t,IO_SPACE_LIMIT , -resource_size (&pcie->io)); +resource_size (&pcie->io) -1 ); }else pcie->realio = pcie->io;
Greg Kroah-Hartman
maintainerPATCH 112/197 dm kcopyd: avoid softlockup in run_complete_job
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: John Pittman <[email protected]>
[ Upstream commit 784c9a29e99eb40b842c29ecf1cc3a79e00fb629 ]
It was reported that softlockups occur when using dm-snapshot ontop of slow (rbd) storage. E.g.:
[ 4047.990647] watchdog: BUG: soft lockup - CPU#10 stuck for 22s! [kworker/10:23:26177] ... [ 4048.034151] Workqueue: kcopyd do_work [dm_mod] [ 4048.034156] RIP: 0010:copy_callback+0x41/0x160 [dm_snapshot] ... [ 4048.034190] Call Trace: [ 4048.034196] ? __chunk_is_tracked+0x70/0x70 [dm_snapshot] [ 4048.034200] run_complete_job+0x5f/0xb0 [dm_mod] [ 4048.034205] process_jobs+0x91/0x220 [dm_mod] [ 4048.034210] ? kcopyd_put_pages+0x40/0x40 [dm_mod] [ 4048.034214] do_work+0x46/0xa0 [dm_mod] [ 4048.034219] process_one_work+0x171/0x370 [ 4048.034221] worker_thread+0x1fc/0x3f0 [ 4048.034224] kthread+0xf8/0x130 [ 4048.034226] ? max_active_store+0x80/0x80 [ 4048.034227] ? kthread_bind+0x10/0x10 [ 4048.034231] ret_from_fork+0x35/0x40 [ 4048.034233] Kernel panic - not syncing: softlockup: hung tasks
Fix this by calling cond_resched() after run_complete_job()'s callout to the dm_kcopyd_notify_fn (which is dm-snap.c:copy_callback in the above trace).
--- drivers/md/dm-kcopyd.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/md/dm-kcopyd.c +2 -0
@@ -487,6 +487,8 @@ static int run_complete_job(struct kcopyif (atomic_dec_and_test (&kc->nr_jobs))wake_up (&kc->destroyq); +cond_resched (); +return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 113/197 staging: comedi: ni_mio_common: fix subdevice flags for PFI subdevice
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ian Abbott <[email protected]>
[ Upstream commit e083926b3e269d4064825dcf2ad50c636fddf8cf ]
The PFI subdevice flags indicate that the subdevice is readable and writeable, but that is only true for the supported "M-series" boards, not the older "E-series" boards. Only set the SDF_READABLE and SDF_WRITABLE subdevice flags for the M-series boards. These two flags are mainly for informational purposes.
--- drivers/staging/comedi/drivers/ni_mio_common.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/staging/comedi/drivers/ni_mio_common.c +2 -1
@@ -5446,11 +5446,11 @@ static int ni_E_init(struct comedi_devic/* Digital I/O (PFI) subdevice */ s = &dev->subdevices[NI_PFI_DIO_SUBDEV ]; s->type =COMEDI_SUBD_DIO ; - s->subdev_flags =SDF_READABLE |SDF_WRITABLE |SDF_INTERNAL ; s->maxdata =1 ;if (devpriv->is_m_series) { s->n_chan =16 ; s->insn_bits = ni_pfi_insn_bits; + s->subdev_flags =SDF_READABLE |SDF_WRITABLE |SDF_INTERNAL ;ni_writew (dev, s->state,NI_M_PFI_DO_REG );for (i =0 ; i <NUM_PFI_OUTPUT_SELECT_REGS ; ++i) {
@@ -5459,6 +5459,7 @@ static int ni_E_init(struct comedi_devic } }else { s->n_chan =10 ; + s->subdev_flags =SDF_INTERNAL ; } s->insn_config = ni_pfi_insn_config;
Greg Kroah-Hartman
maintainerPATCH 114/197 ASoC: rt5677: Fix initialization of rt5677_of_match.data
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Matthias Kaehlcke <[email protected]>
[ Upstream commit f861e3e28a3016a2064d9f600eaa92a530b732b4 ]
The driver expects to find the device id in rt5677_of_match.data, however it is currently assigned to rt5677_of_match.type. Fix this.
The problem was found with the help of clang:
sound/soc/codecs/rt5677.c:5010:36: warning: expression which evaluates to zero treated as a null pointer constant of type 'const void *' [-Wnon-literal-null-conversion]
{ .compatible = "realtek,rt5677", RT5677 },
^~~~~~--- sound/soc/codecs/rt5677.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/codecs/rt5677.c +1 -1
@@ -5007,7 +5007,7 @@ static const struct regmap_config rt5677 };static const struct of_device_id rt5677_of_match[] = { - { .compatible ="realtek,rt5677" ,RT5677 }, + { .compatible ="realtek,rt5677" , .data = (const void *)RT5677 }, { } };MODULE_DEVICE_TABLE (of, rt5677_of_match);
Greg Kroah-Hartman
maintainerPATCH 115/197 iommu/omap: Fix cache flushes on L2 table entries
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ralf Goebel <[email protected]>
[ Upstream commit 04c532a1cdc7e423656c07937aa4b5c1c2b064f9 ]
The base address used for DMA operations on the second-level table did incorrectly include the offset for the table entry. The offset was then added again which lead to incorrect behavior.
Operations on the L1 table are not affected.
The calculation of the base address is changed to point to the beginning of the L2 table.
--- drivers/iommu/omap-iommu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/iommu/omap-iommu.c +2 -2
@@ -550,7 +550,7 @@ static u32 *iopte_alloc(struct omap_iomm pte_ready: iopte =iopte_offset (iopgd, da); - *pt_dma =virt_to_phys (iopte); + *pt_dma =iopgd_page_paddr (iopgd);dev_vdbg (obj->dev,"%s: da:%08x pgd:%p *pgd:%08x pte:%p *pte:%08x\n" , __func__, da, iopgd, *iopgd, iopte, *iopte);
@@ -738,7 +738,7 @@ static size_t iopgtable_clear_entry_core } bytes *= nent;memset (iopte,0 , nent *sizeof (*iopte)); - pt_dma =virt_to_phys (iopte); + pt_dma =iopgd_page_paddr (iopgd);flush_iopte_range (obj->dev, pt_dma, pt_offset, nent);/*
Greg Kroah-Hartman
maintainerPATCH 116/197 selftests/powerpc: Kill child processes on SIGINT
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Breno Leitao <[email protected]>
[ Upstream commit 7c27a26e1ed5a7dd709aa19685d2c98f64e1cf0c ]
There are some powerpc selftests, as tm/tm-unavailable, that run for a long period (>120 seconds), and if it is interrupted, as pressing CRTL-C (SIGINT), the foreground process (harness) dies but the child process and threads continue to execute (with PPID = 1 now) in background.
In this case, you'd think the whole test exited, but there are remaining threads and processes being executed in background. Sometimes these zombies processes are doing annoying things, as consuming the whole CPU or dumping things to STDOUT.
This patch fixes this problem by attaching an empty signal handler to SIGINT in the harness process. This handler will interrupt (EINTR) the parent process waitpid() call, letting the code to follow through the normal flow, which will kill all the processes in the child process group.
This patch also fixes a typo.
--- tools/testing/selftests/powerpc/harness.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-)
tools/testing/selftests/powerpc/harness.c +12 -6
@@ -85,13 +85,13 @@ wait:return status; } -static void alarm_handler (int signum) +static void sig_handler (int signum) { -/* Jut wake us up from waitpid */ +/* Just wake us up from waitpid */ } -static struct sigaction alarm_action = { - .sa_handler = alarm_handler, +static struct sigaction sig_action = { + .sa_handler = sig_handler, };void test_harness_set_timeout (uint64_t time)
@@ -106,8 +106,14 @@ int test_harness(int (test_function)(voitest_start (name);test_set_git_version (GIT_VERSION ); -if (sigaction (SIGALRM , &alarm_action,NULL )) { -perror ("sigaction" ); +if (sigaction (SIGINT , &sig_action,NULL )) { +perror ("sigaction (sigint)" ); +test_error (name); +return 1 ; + } + +if (sigaction (SIGALRM , &sig_action,NULL )) { +perror ("sigaction (sigalrm)" );test_error (name);return 1 ; }
Greg Kroah-Hartman
maintainerPATCH 117/197 selinux: cleanup dentry and inodes on error in selinuxfs
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: nixiaoming <[email protected]>
[ Upstream commit 7e4237faa7213c1cc1d0aa65a44c67ba4729ce9f ]
If the resource requested by d_alloc_name is not added to the linked list through d_add, then dput needs to be called to release the subsequent abnormal branch to avoid resource leakage.
Add missing dput to selinuxfs.c
--- security/selinux/selinuxfs.c | 33 +++++++++++++++++++++++++-------- 1 file changed, 25 insertions(+), 8 deletions(-)
security/selinux/selinuxfs.c +25 -8
@@ -1365,13 +1365,18 @@ static int sel_make_bools(struct selinux ret = -ENOMEM ; inode =sel_make_inode (dir->d_sb,S_IFREG |S_IRUGO |S_IWUSR ); -if (!inode) +if (!inode) { +dput (dentry); goto out; + } ret = -ENAMETOOLONG ; len =snprintf (page,PAGE_SIZE ,"/%s/%s" ,BOOL_DIR_NAME , names[i]); -if (len >=PAGE_SIZE ) +if (len >=PAGE_SIZE ) { +dput (dentry); +iput (inode); goto out; + } isec = (struct inode_security_struct *)inode->i_security; ret = security_genfs_sid(fsi->state, "selinuxfs", page,
@@ -1586,8 +1591,10 @@ static int sel_make_avc_files(struct denreturn -ENOMEM ; inode =sel_make_inode (dir->d_sb,S_IFREG |files[i].mode); -if (!inode) +if (!inode) { +dput (dentry);return -ENOMEM ; + } inode->i_fop = files[i].ops; inode->i_ino = ++fsi->last_ino;
@@ -1632,8 +1639,10 @@ static int sel_make_initcon_files(structreturn -ENOMEM ; inode =sel_make_inode (dir->d_sb,S_IFREG |S_IRUGO ); -if (!inode) +if (!inode) { +dput (dentry);return -ENOMEM ; + } inode->i_fop = &sel_initcon_ops; inode->i_ino = i|SEL_INITCON_INO_OFFSET ;
@@ -1733,8 +1742,10 @@ static int sel_make_perm_files(char *obj rc = -ENOMEM ; inode =sel_make_inode (dir->d_sb,S_IFREG |S_IRUGO ); -if (!inode) +if (!inode) { +dput (dentry); goto out; + } inode->i_fop = &sel_perm_ops;/* i+1 since perm values are 1-indexed */
@@ -1763,8 +1774,10 @@ static int sel_make_class_dir_entries(chreturn -ENOMEM ; inode =sel_make_inode (dir->d_sb,S_IFREG |S_IRUGO ); -if (!inode) +if (!inode) { +dput (dentry);return -ENOMEM ; + } inode->i_fop = &sel_class_ops; inode->i_ino =sel_class_to_ino (index);
@@ -1838,8 +1851,10 @@ static int sel_make_policycap(struct selreturn -ENOMEM ; inode =sel_make_inode (fsi->sb,S_IFREG |0444 ); -if (inode ==NULL ) +if (inode ==NULL ) { +dput (dentry);return -ENOMEM ; + } inode->i_fop = &sel_policycap_ops; inode->i_ino = iter |SEL_POLICYCAP_INO_OFFSET ;
@@ -1932,8 +1947,10 @@ static int sel_fill_super(struct super_b ret = -ENOMEM ; inode =sel_make_inode (sb,S_IFCHR |S_IRUGO |S_IWUGO ); -if (!inode) +if (!inode) { +dput (dentry); goto err; + } inode->i_ino = ++fsi->last_ino; isec = (struct inode_security_struct *)inode->i_security;
Greg Kroah-Hartman
maintainerPATCH 118/197 RDS: IB: fix passing zero to ERR_PTR() warning
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: YueHaibing <[email protected]>
[ Upstream commit 5941923da29e84bc9e2a1abb2c14fffaf8d71e2f ]
Fix a static code checker warning:
net/rds/ib_frmr.c:82 rds_ib_alloc_frmr() warn: passing zero to 'ERR_PTR'
The error path for ib_alloc_mr failure should set err to PTR_ERR.
--- net/rds/ib_frmr.c | 1 + 1 file changed, 1 insertion(+)
net/rds/ib_frmr.c +1 -0
@@ -61,6 +61,7 @@ static struct rds_ib_mr *rds_ib_alloc_fr pool->fmr_attr.max_pages);if (IS_ERR (frmr->mr)) {pr_warn ("RDS/IB: %s failed to allocate MR" , __func__); + err =PTR_ERR (frmr->mr); goto out_no_cigar; }
Greg Kroah-Hartman
maintainerPATCH 119/197 cfq: Suppress compiler warnings about comparisons
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Bart Van Assche <[email protected]>
[ Upstream commit f7ecb1b109da1006a08d5675debe60990e824432 ]
This patch does not change any functionality but avoids that gcc reports the following warnings when building with W=1:
block/cfq-iosched.c: In function ?cfq_back_seek_max_store?:block/cfq-iosched.c:4741:13: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (__data < (MIN)) \
^block/cfq-iosched.c:4756:1: note: in expansion of macro ?STORE_FUNCTION?
STORE_FUNCTION(cfq_back_seek_max_store, &cfqd->cfq_back_max, 0, UINT_MAX, 0); ^~~~~~~~~~~~~~block/cfq-iosched.c: In function ?cfq_slice_idle_store?:block/cfq-iosched.c:4741:13: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (__data < (MIN)) \
^block/cfq-iosched.c:4759:1: note: in expansion of macro ?STORE_FUNCTION?
STORE_FUNCTION(cfq_slice_idle_store, &cfqd->cfq_slice_idle, 0, UINT_MAX, 1); ^~~~~~~~~~~~~~block/cfq-iosched.c: In function ?cfq_group_idle_store?:block/cfq-iosched.c:4741:13: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (__data < (MIN)) \
^block/cfq-iosched.c:4760:1: note: in expansion of macro ?STORE_FUNCTION?
STORE_FUNCTION(cfq_group_idle_store, &cfqd->cfq_group_idle, 0, UINT_MAX, 1); ^~~~~~~~~~~~~~block/cfq-iosched.c: In function ?cfq_low_latency_store?:block/cfq-iosched.c:4741:13: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (__data < (MIN)) \
^block/cfq-iosched.c:4765:1: note: in expansion of macro ?STORE_FUNCTION?
STORE_FUNCTION(cfq_low_latency_store, &cfqd->cfq_latency, 0, 1, 0);
^~~~~~~~~~~~~~block/cfq-iosched.c: In function ?cfq_slice_idle_us_store?:block/cfq-iosched.c:4775:13: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (__data < (MIN)) \
^block/cfq-iosched.c:4782:1: note: in expansion of macro ?USEC_STORE_FUNCTION?
USEC_STORE_FUNCTION(cfq_slice_idle_us_store, &cfqd->cfq_slice_idle, 0, UINT_MAX); ^~~~~~~~~~~~~~~~~~~block/cfq-iosched.c: In function ?cfq_group_idle_us_store?:block/cfq-iosched.c:4775:13: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (__data < (MIN)) \
^block/cfq-iosched.c:4783:1: note: in expansion of macro ?USEC_STORE_FUNCTION?
USEC_STORE_FUNCTION(cfq_group_idle_us_store, &cfqd->cfq_group_idle, 0, UINT_MAX); ^~~~~~~~~~~~~~~~~~~
--- block/cfq-iosched.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-)
block/cfq-iosched.c +12 -10
@@ -4735,12 +4735,13 @@ USEC_SHOW_FUNCTION(cfq_target_latency_usstatic ssize_t __FUNC (struct elevator_queue *e,const char *page,size_t count) \ { \struct cfq_data *cfqd = e->elevator_data; \ -unsigned int __data; \ +unsigned int __data, __min = (MIN ), __max = (MAX ); \ + \cfq_var_store (&__data, (page)); \ -if (__data < (MIN )) \ - __data = (MIN ); \ -else if (__data > (MAX )) \ - __data = (MAX ); \ +if (__data < __min) \ + __data = __min; \ +else if (__data > __max) \ + __data = __max; \if (__CONV) \ *(__PTR) = (u64 )__data *NSEC_PER_MSEC ; \else \
@@ -4769,12 +4770,13 @@ STORE_FUNCTION(cfq_target_latency_store,static ssize_t __FUNC (struct elevator_queue *e,const char *page,size_t count) \ { \struct cfq_data *cfqd = e->elevator_data; \ -unsigned int __data; \ +unsigned int __data, __min = (MIN ), __max = (MAX ); \ + \cfq_var_store (&__data, (page)); \ -if (__data < (MIN )) \ - __data = (MIN ); \ -else if (__data > (MAX )) \ - __data = (MAX ); \ +if (__data < __min) \ + __data = __min; \ +else if (__data > __max) \ + __data = __max; \ *(__PTR) = (u64 )__data *NSEC_PER_USEC ; \return count; \ }
Greg Kroah-Hartman
maintainerPATCH 120/197 smb3: fix reset of bytes read and written stats
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steve French <[email protected]>
[ Upstream commit c281bc0c7412308c7ec0888904f7c99353da4796 ]
echo 0 > /proc/fs/cifs/Stats is supposed to reset the stats but there were four (see example below) that were not reset (bytes read and witten, total vfs ops and max ops at one time).
...
0 session 0 share reconnects
Total vfs operations: 100 maximum at one time: 2
1) \\localhost\test
SMBs: 0
Bytes read: 502092 Bytes written: 31457286
TreeConnects: 0 total 0 failed
TreeDisconnects: 0 total 0 failed
...
This patch fixes cifs_stats_proc_write to properly reset those four.
--- fs/cifs/cifs_debug.c | 8 ++++++++ 1 file changed, 8 insertions(+)
fs/cifs/cifs_debug.c +8 -0
@@ -383,6 +383,10 @@ static ssize_t cifs_stats_proc_write(stratomic_set (&totBufAllocCount,0 );atomic_set (&totSmBufAllocCount,0 );#endif /* CONFIG_CIFS_STATS2 */ +spin_lock (&GlobalMid_Lock); + GlobalMaxActiveXid =0 ; + GlobalCurrentXid =0 ; +spin_unlock (&GlobalMid_Lock);spin_lock (&cifs_tcp_ses_lock);list_for_each (tmp1, &cifs_tcp_ses_list) { server = list_entry(tmp1, struct TCP_Server_Info,
@@ -395,6 +399,10 @@ static ssize_t cifs_stats_proc_write(str struct cifs_tcon, tcon_list);atomic_set (&tcon->num_smbs_sent,0 ); +spin_lock (&tcon->stat_lock); + tcon->bytes_read =0 ; + tcon->bytes_written =0 ; +spin_unlock (&tcon->stat_lock);if (server->ops->clear_stats) server->ops->clear_stats (tcon); }
Greg Kroah-Hartman
maintainerPATCH 121/197 CIFS: fix memory leak and remove dead code
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aurelien Aptel <[email protected]>
[ Upstream commit 256b4c3f03d77d8c0dc69e3a6ceb3afd0d1810bd ]
also fixes error code in smb311_posix_mkdir() (where the error assignment needs to go before the goto) a typo that Dan Carpenter and Paulo and Gustavo pointed out.
--- fs/cifs/smb2pdu.c | 101 ++++++++++++++++++++++++++---------------------------- 1 file changed, 50 insertions(+), 51 deletions(-)
fs/cifs/smb2pdu.c +50 -51
@@ -1928,7 +1928,7 @@ int smb311_posix_mkdir(const unsigned in {struct smb_rqst rqst;struct smb2_create_req *req; -struct smb2_create_rsp *rsp; +struct smb2_create_rsp *rsp =NULL ;struct TCP_Server_Info *server;struct cifs_ses *ses = tcon->ses;struct kvec iov[3 ];/* make sure at least one for each open context */
@@ -1943,27 +1943,31 @@ int smb311_posix_mkdir(const unsigned inchar *pc_buf =NULL ;int flags =0 ;unsigned int total_len; -__le16 *path =cifs_convert_path_to_utf16 (full_path, cifs_sb); - -if (!path) -return -ENOMEM ; +__le16 *utf16_path =NULL ;cifs_dbg (FYI ,"mkdir\n" ); +/* resource #1: path allocation */ + utf16_path =cifs_convert_path_to_utf16 (full_path, cifs_sb); +if (!utf16_path) +return -ENOMEM ; +if (ses && (ses->server)) server = ses->server; -else -return -EIO ; +else { + rc = -EIO ; + goto err_free_path; + } +/* resource #2: request */ rc =smb2_plain_req_init (SMB2_CREATE , tcon, (void **) &req, &total_len); -if (rc) -return rc; + goto err_free_path; +if (smb3_encryption_required (tcon)) flags |=CIFS_TRANSFORM_REQ ; - req->ImpersonationLevel =IL_IMPERSONATION ; req->DesiredAccess =cpu_to_le32 (FILE_WRITE_ATTRIBUTES );/* File attributes ignored on open (used in create though) */
@@ -1992,50 +1996,44 @@ int smb311_posix_mkdir(const unsigned in req->sync_hdr.Flags |=SMB2_FLAGS_DFS_OPERATIONS ; rc =alloc_path_with_tree_prefix (©_path, ©_size, &name_len, - tcon->treeName, path); -if (rc) { -cifs_small_buf_release (req); -return rc; - } + tcon->treeName, utf16_path); +if (rc) + goto err_free_req; + req->NameLength =cpu_to_le16 (name_len *2 ); uni_path_len = copy_size; - path = copy_path; +/* free before overwriting resource */ +kfree (utf16_path); + utf16_path = copy_path; }else { - uni_path_len = (2 *UniStrnlen ((wchar_t *)path,PATH_MAX )) +2 ; + uni_path_len = (2 *UniStrnlen ((wchar_t *)utf16_path,PATH_MAX )) +2 ;/* MUST set path len (NameLength) to 0 opening root of share */ req->NameLength =cpu_to_le16 (uni_path_len -2 );if (uni_path_len %8 !=0 ) { copy_size =roundup (uni_path_len,8 ); copy_path =kzalloc (copy_size,GFP_KERNEL );if (!copy_path) { -cifs_small_buf_release (req); -return -ENOMEM ; + rc = -ENOMEM ; + goto err_free_req; } -memcpy ((char *)copy_path, (const char *)path, +memcpy ((char *)copy_path, (const char *)utf16_path, uni_path_len); uni_path_len = copy_size; - path = copy_path; +/* free before overwriting resource */ +kfree (utf16_path); + utf16_path = copy_path; } } iov[1 ].iov_len = uni_path_len; - iov[1 ].iov_base = path; + iov[1 ].iov_base = utf16_path; req->RequestedOplockLevel =SMB2_OPLOCK_LEVEL_NONE ;if (tcon->posix_extensions) { -if (n_iov >2 ) { -struct create_context *ccontext = - (struct create_context *)iov[n_iov-1 ].iov_base; - ccontext->Next = -cpu_to_le32 (iov[n_iov-1 ].iov_len); - } - +/* resource #3: posix buf */ rc =add_posix_context (iov, &n_iov, mode); -if (rc) { -cifs_small_buf_release (req); -kfree (copy_path); -return rc; - } +if (rc) + goto err_free_req; pc_buf = iov[n_iov-1 ].iov_base; }
@@ -2044,32 +2042,33 @@ int smb311_posix_mkdir(const unsigned in rqst.rq_iov = iov; rqst.rq_nvec = n_iov; - rc =cifs_send_recv (xid, ses, &rqst, &resp_buftype, flags, - &rsp_iov); - -cifs_small_buf_release (req); - rsp = (struct smb2_create_rsp *)rsp_iov.iov_base; - -if (rc !=0 ) { +/* resource #4: response buffer */ + rc =cifs_send_recv (xid, ses, &rqst, &resp_buftype, flags, &rsp_iov); +if (rc) {cifs_stats_fail_inc (tcon,SMB2_CREATE_HE );trace_smb3_posix_mkdir_err (xid, tcon->tid, ses->Suid, -CREATE_NOT_FILE ,FILE_WRITE_ATTRIBUTES , rc); - goto smb311_mkdir_exit; - }else -trace_smb3_posix_mkdir_done (xid, rsp->PersistentFileId, tcon->tid, - ses->Suid,CREATE_NOT_FILE , -FILE_WRITE_ATTRIBUTES ); +CREATE_NOT_FILE , +FILE_WRITE_ATTRIBUTES , rc); + goto err_free_rsp_buf; + } + + rsp = (struct smb2_create_rsp *)rsp_iov.iov_base; +trace_smb3_posix_mkdir_done (xid, rsp->PersistentFileId, tcon->tid, + ses->Suid,CREATE_NOT_FILE , +FILE_WRITE_ATTRIBUTES );SMB2_close (xid, tcon, rsp->PersistentFileId, rsp->VolatileFileId);/* Eventually save off posix specific response info and timestaps */ -smb311_mkdir_exit: -kfree (copy_path); -kfree (pc_buf); +err_free_rsp_buf:free_rsp_buf (resp_buftype, rsp); +kfree (pc_buf); +err_free_req: +cifs_small_buf_release (req); +err_free_path: +kfree (utf16_path); return rc; - }#endif /* SMB311 */
Greg Kroah-Hartman
maintainerPATCH 122/197 SMB3: Number of requests sent should be displayed for SMB3 not just CIFS
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steve French <[email protected]>
[ Upstream commit 289131e1f1e6ad8c661ec05e176b8f0915672059 ]
For SMB2/SMB3 the number of requests sent was not displayed in /proc/fs/cifs/Stats unless CONFIG_CIFS_STATS2 was enabled (only number of failed requests displayed). As with earlier dialects, we should be displaying these counters if CONFIG_CIFS_STATS is enabled. They
are important for debugging.
e.g. when you cat /proc/fs/cifs/Stats (before the patch) Resources in use
CIFS Session: 1
Share (unique mount targets): 2
SMB Request/Response Buffer: 1 Pool size: 5
SMB Small Req/Resp Buffer: 1 Pool size: 30 Operations (MIDs): 0
0 session 0 share reconnects
Total vfs operations: 690 maximum at one time: 2
1) \\localhost\test
SMBs: 975
Negotiates: 0 sent 0 failed
SessionSetups: 0 sent 0 failed
Logoffs: 0 sent 0 failed
TreeConnects: 0 sent 0 failed
TreeDisconnects: 0 sent 0 failed
Creates: 0 sent 2 failed
Closes: 0 sent 0 failed
Flushes: 0 sent 0 failed
Reads: 0 sent 0 failed
Writes: 0 sent 0 failed
Locks: 0 sent 0 failed
IOCTLs: 0 sent 1 failed
Cancels: 0 sent 0 failed
Echos: 0 sent 0 failed
QueryDirectories: 0 sent 63 failed
--- fs/cifs/smb2pdu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/cifs/smb2pdu.c +1 -1
@@ -360,7 +360,7 @@ smb2_plain_req_init(__le16 smb2_command, total_len);if (tcon !=NULL ) { -#ifdef CONFIG_CIFS_STATS2 +#ifdef CONFIG_CIFS_STATS uint16_t com_code =le16_to_cpu (smb2_command);cifs_stats_inc (&tcon->stats.smb2_stats.smb2_com_sent[com_code]);#endif
Greg Kroah-Hartman
maintainerPATCH 123/197 smb3: if server does not support posix do not allow posix mount option
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steve French <[email protected]>
[ Upstream commit 8505c8bfd85a260c9dc5c47e15bd8c5357fcbcd2 ]
If user specifies "posix" on an SMB3.11 mount, then fail the mount if server does not return the POSIX negotiate context indicating support for posix.
--- fs/cifs/connect.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
fs/cifs/connect.c +6 -2
@@ -3031,11 +3031,15 @@ cifs_get_tcon(struct cifs_ses *ses, stru }#ifdef CONFIG_CIFS_SMB311 -if ((volume_info->linux_ext) && (ses->server->posix_ext_supported)) { -if (ses->server->vals->protocol_id ==SMB311_PROT_ID ) { +if (volume_info->linux_ext) { +if (ses->server->posix_ext_supported) { tcon->posix_extensions = true;printk_once (KERN_WARNING "SMB3.11 POSIX Extensions are experimental\n" ); + }else { +cifs_dbg (VFS ,"Server does not support mounting with posix SMB3.11 extensions.\n" ); + rc = -EOPNOTSUPP ; + goto out_fail; } }#endif /* 311 */
Greg Kroah-Hartman
maintainerPATCH 124/197 powerpc/platforms/85xx: fix t1042rdb_diu.c build errors & warning
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Randy Dunlap <[email protected]>
[ Upstream commit f5daf77a55ef0e695cc90c440ed6503073ac5e07 ]
Fix build errors and warnings in t1042rdb_diu.c by adding header files and MODULE_LICENSE().
../arch/powerpc/platforms/85xx/t1042rdb_diu.c:152:1: warning: data definition has no type or storage class
early_initcall(t1042rdb_diu_init);../arch/powerpc/platforms/85xx/t1042rdb_diu.c:152:1: error: type defaults to 'int' in declaration of 'early_initcall' [-Werror=implicit-int] ../arch/powerpc/platforms/85xx/t1042rdb_diu.c:152:1: warning: parameter names (without types) in function declaration
and
WARNING: modpost: missing MODULE_LICENSE() in arch/powerpc/platforms/85xx/t1042rdb_diu.o
--- arch/powerpc/platforms/85xx/t1042rdb_diu.c | 4 ++++ 1 file changed, 4 insertions(+)
arch/powerpc/platforms/85xx/t1042rdb_diu.c +4 -0
@@ -9,8 +9,10 @@* option) any later version. */ +#include <linux/init.h> #include <linux/io.h> #include <linux/kernel.h> +#include <linux/module.h> #include <linux/of.h> #include <linux/of_address.h>
@@ -150,3 +152,5 @@ static int __init t1042rdb_diu_init(void }early_initcall (t1042rdb_diu_init); + +MODULE_LICENSE ("GPL" );
Greg Kroah-Hartman
maintainerPATCH 125/197 powerpc/64s: Make rfi_flush_fallback a little more robust
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michael Ellerman <[email protected]>
[ Upstream commit 78ee9946371f5848ddfc88ab1a43867df8f17d83 ]
Because rfi_flush_fallback runs immediately before the return to userspace it currently runs with the user r1 (stack pointer). This means if we oops in there we will report a bad kernel stack pointer in the exception entry path, eg:
Bad kernel stack pointer 7ffff7150e40 at c0000000000023b4
Oops: Bad kernel stack pointer, sig: 6 [#1] LE SMP NR_CPUS=32 NUMA PowerNV Modules linked in: CPU: 0 PID: 1246 Comm: klogd Not tainted 4.18.0-rc2-gcc-7.3.1-00175-g0443f8a69ba3 #7 NIP: c0000000000023b4 LR: 0000000010053e00 CTR: 0000000000000040 REGS: c0000000fffe7d40 TRAP: 4100 Not tainted (4.18.0-rc2-gcc-7.3.1-00175-g0443f8a69ba3) MSR: 9000000002803031 <SF,HV,VEC,VSX,FP,ME,IR,DR,LE> CR: 44000442 XER: 20000000 CFAR: c00000000000bac8 IRQMASK: c0000000f1e66a80 GPR00: 0000000002000000 00007ffff7150e40 00007fff93a99900 0000000000000020 ... NIP [c0000000000023b4] rfi_flush_fallback+0x34/0x80 LR [0000000010053e00] 0x10053e00
Although the NIP tells us where we were, and the TRAP number tells us what happened, it would still be nicer if we could report the actual exception rather than barfing about the stack pointer.
We an do that fairly simply by loading the kernel stack pointer on entry and restoring the user value before returning. That way we see a regular oops such as:
Unrecoverable exception 4100 at c00000000000239c
Oops: Unrecoverable exception, sig: 6 [#1] LE SMP NR_CPUS=32 NUMA PowerNV Modules linked in: CPU: 0 PID: 1251 Comm: klogd Not tainted 4.18.0-rc3-gcc-7.3.1-00097-g4ebfcac65acd-dirty #40 NIP: c00000000000239c LR: 0000000010053e00 CTR: 0000000000000040 REGS: c0000000f1e17bb0 TRAP: 4100 Not tainted (4.18.0-rc3-gcc-7.3.1-00097-g4ebfcac65acd-dirty) MSR: 9000000002803031 <SF,HV,VEC,VSX,FP,ME,IR,DR,LE> CR: 44000442 XER: 20000000 CFAR: c00000000000bac8 IRQMASK: 0 ... NIP [c00000000000239c] rfi_flush_fallback+0x3c/0x80 LR [0000000010053e00] 0x10053e00 Call Trace: [c0000000f1e17e30] [c00000000000b9e4] system_call+0x5c/0x70 (unreliable)
Note this shouldn't make the kernel stack pointer vulnerable to a meltdown attack, because it should be flushed from the cache before we return to userspace. The user r1 value will be in the cache, because we load it in the return path, but that is harmless.
--- arch/powerpc/kernel/exceptions-64s.S | 6 ++++++ 1 file changed, 6 insertions(+)
arch/powerpc/kernel/exceptions-64s.S +6 -0
@@ -1526,6 +1526,8 @@ TRAMP_REAL_BEGIN(stf_barrier_fallback) TRAMP_REAL_BEGIN(rfi_flush_fallback) SET_SCRATCH0(r13); GET_PACA(r13); + std r1,PACA_EXRFI+EX_R12(r13) + ld r1,PACAKSAVE(r13) std r9,PACA_EXRFI+EX_R9(r13) std r10,PACA_EXRFI+EX_R10(r13) std r11,PACA_EXRFI+EX_R11(r13)
@@ -1560,12 +1562,15 @@ TRAMP_REAL_BEGIN(rfi_flush_fallback) ld r9,PACA_EXRFI+EX_R9(r13) ld r10,PACA_EXRFI+EX_R10(r13) ld r11,PACA_EXRFI+EX_R11(r13) + ld r1,PACA_EXRFI+EX_R12(r13) GET_SCRATCH0(r13); rfid TRAMP_REAL_BEGIN(hrfi_flush_fallback) SET_SCRATCH0(r13); GET_PACA(r13); + std r1,PACA_EXRFI+EX_R12(r13) + ld r1,PACAKSAVE(r13) std r9,PACA_EXRFI+EX_R9(r13) std r10,PACA_EXRFI+EX_R10(r13) std r11,PACA_EXRFI+EX_R11(r13)
@@ -1600,6 +1605,7 @@ TRAMP_REAL_BEGIN(hrfi_flush_fallback) ld r9,PACA_EXRFI+EX_R9(r13) ld r10,PACA_EXRFI+EX_R10(r13) ld r11,PACA_EXRFI+EX_R11(r13) + ld r1,PACA_EXRFI+EX_R12(r13) GET_SCRATCH0(r13); hrfid
Greg Kroah-Hartman
maintainerPATCH 126/197 um: fix parallel building with O= option
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Masahiro Yamada <[email protected]>
[ Upstream commit 13d3d01e26b942ada7cfced68ccb6db49597874a ]
Randy Dunlap reports UML occasionally fails to build with -j<N> and O=<builddir> options.
make[1]: Entering directory '/home/rdunlap/mmotm-2018-0802-1529/UM64'
UPD include/generated/uapi/linux/version.h
WRAP arch/x86/include/generated/asm/dma-contiguous.h
WRAP arch/x86/include/generated/asm/export.h
WRAP arch/x86/include/generated/asm/early_ioremap.h
WRAP arch/x86/include/generated/asm/mcs_spinlock.h
WRAP arch/x86/include/generated/asm/mm-arch-hooks.h
WRAP arch/x86/include/generated/uapi/asm/bpf_perf_event.h
WRAP arch/x86/include/generated/uapi/asm/poll.h
GEN ./Makefile make[2]: *** No rule to make target 'archheaders'. Stop. arch/um/Makefile:119: recipe for target 'archheaders' failed make[1]: *** [archheaders] Error 2
make[1]: *** Waiting for unfinished jobs....
UPD include/config/kernel.release
make[1]: *** wait: No child processes. Stop. Makefile:146: recipe for target 'sub-make' failed
make: *** [sub-make] Error 2
The cause of the problem is the use of '$(MAKE) KBUILD_SRC=', which recurses to the top Makefile via the $(objtree)/Makefile generated by scripts/mkmakefile.
When you run "make -j<N> O=<builddir> ARCH=um", Make can execute 'archheaders' and 'outputmakefile' targets simultaneously because there is no dependency between them.
If it happens,
$(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders
... tries to run $(objtree)/Makefile that is being updated.
The correct way for the recursion is
$(Q)$(MAKE) -f $(srctree)/Makefile ARCH=$(HEADER_ARCH) archheaders
..., which does not rely on the generated Makefile.
--- arch/um/Makefile | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
arch/um/Makefile +1 -2
@@ -122,8 +122,7 @@ archheaders: $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.asm-generic \ kbuild-file=$(HOST_DIR)/include/uapi/asm/Kbuild \ obj=$(HOST_DIR)/include/generated/uapi/asm - $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders - + $(Q)$(MAKE) -f $(srctree)/Makefile ARCH=$(HEADER_ARCH) archheaders archprepare: include/generated/user_constants.h
Greg Kroah-Hartman
maintainerPATCH 127/197 powerpc/pseries: Avoid using the size greater than RTAS_ERROR_LOG_MAX.
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mahesh Salgaonkar <[email protected]>
[ Upstream commit 74e96bf44f430cf7a01de19ba6cf49b361cdfd6e ]
The global mce data buffer that used to copy rtas error log is of 2048 (RTAS_ERROR_LOG_MAX) bytes in size. Before the copy we read extended_log_length from rtas error log header, then use max of extended_log_length and RTAS_ERROR_LOG_MAX as a size of data to be copied. Ideally the platform (phyp) will never send extended error log with size > 2048. But if that happens, then we have a risk of buffer overrun and corruption. Fix this by using min_t instead.
--- arch/powerpc/platforms/pseries/ras.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/pseries/ras.c +1 -1
@@ -371,7 +371,7 @@ static struct rtas_error_log *fwnmi_get_int len, error_log_length; error_log_length =8 +rtas_error_extended_log_length (h); - len =max_t (int, error_log_length,RTAS_ERROR_LOG_MAX ); + len =min_t (int, error_log_length,RTAS_ERROR_LOG_MAX );memset (global_mce_data_buf,0 ,RTAS_ERROR_LOG_MAX );memcpy (global_mce_data_buf, h, len); errhdr = (struct rtas_error_log *)global_mce_data_buf;
Greg Kroah-Hartman
maintainerPATCH 128/197 clk: rockchip: Add pclk_rkpwm_pmu to PMU critical clocks in rk3399
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Levin Du <[email protected]>
[ Upstream commit 640332d1a089909df08bc9f3e42888a2019c66e2 ]
PWM2 is commonly used to control voltage of PWM regulator of VDD_LOG in RK3399. On the Firefly-RK3399 board, PWM2 outputs 40 KHz square wave from power on and the VDD_LOG is about 0.9V. When the kernel boots normally into the system, the PWM2 keeps outputing PWM signal.
But the kernel hangs randomly after "Starting kernel ..." line on that board. When it happens, PWM2 outputs high level which causes VDD_LOG drops to 0.4V below the normal operating voltage.
By adding "pclk_rkpwm_pmu" to the rk3399_pmucru_critical_clocks array, PWM clock is ensured to be prepared at startup and the PWM2 output is normal. After repeated tests, the early boot hang is gone.
This patch works on both Firefly-RK3399 and ROC-RK3399-PC boards.
--- drivers/clk/rockchip/clk-rk3399.c | 1 + 1 file changed, 1 insertion(+)
drivers/clk/rockchip/clk-rk3399.c +1 -0
@@ -1523,6 +1523,7 @@ static const char *const rk3399_pmucru_c"pclk_pmu_src" ,"fclk_cm0s_src_pmu" ,"clk_timer_src_pmu" , + "pclk_rkpwm_pmu", };static void __init rk3399_clk_init (struct device_node *np)
Greg Kroah-Hartman
maintainerPATCH 129/197 drm/amd/display: Read back max backlight value at boot
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Francis <[email protected]>
[ Upstream commit 53a53f8687faf492df2644d8c18ff0217fc18730 ]
[Why]
If there is no program explicitly setting the backlight brightness (for example, during a minimal install of linux), the hardware defaults to maximum brightness but the backlight_device defaults to 0 value. Thus, settings displays the wrong brightness value.
[How]
When creating the backlight device, set brightness to max
--- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +1 -0
@@ -1334,6 +1334,7 @@ amdgpu_dm_register_backlight_device(strustruct backlight_properties props = {0 }; props.max_brightness =AMDGPU_MAX_BL_LEVEL ; + props.brightness =AMDGPU_MAX_BL_LEVEL ; props.type =BACKLIGHT_RAW ; snprintf(bl_name, sizeof(bl_name), "amdgpu_bl%d",
Greg Kroah-Hartman
maintainerPATCH 130/197 KVM: vmx: track host_state.loaded using a loaded_vmcs pointer
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson <[email protected]>
[ Upstream commit bd9966de4e14fb559e89a06f7f5c9aab2cc028b9 ]
Using 'struct loaded_vmcs*' to track whether the CPU registers contain host or guest state kills two birds with one stone.
1. The (effective) boolean host_state.loaded is poorly named. It does not track whether or not host state is loaded into the CPU registers (which most readers would expect), but rather tracks if host state has been saved AND guest state is loaded.
2. Using a loaded_vmcs pointer provides a more robust framework for the optimized guest/host state switching, especially when consideration per-VMCS enhancements. To that end, WARN_ONCE if we try to switch to host state with a different VMCS than was last used to save host state.
Resolve an occurrence of the new WARN by setting loaded_vmcs after the call to vmx_vcpu_put() in vmx_switch_vmcs().
--- arch/x86/kvm/vmx.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-)
arch/x86/kvm/vmx.c +15 -7
@@ -939,17 +939,21 @@ struct vcpu_vmx {/* * loaded_vmcs points to the VMCS currently used in this vcpu. For a * non-nested (L1) guest, it always points to vmcs01. For a nested -* guest (L2), it points to a different VMCS. +* guest (L2), it points to a different VMCS. loaded_cpu_state points +* to the VMCS whose state is loaded into the CPU registers that only +* need to be switched when transitioning to/from the kernel; a NULL +* value indicates that host state is loaded. */ struct loaded_vmcs vmcs01;struct loaded_vmcs *loaded_vmcs; +struct loaded_vmcs *loaded_cpu_state;bool __launched;/* temporary, used in vmx_vcpu_run */ struct msr_autoload {struct vmx_msrs guest;struct vmx_msrs host; } msr_autoload; +struct { -int loaded;u16 fs_sel, gs_sel, ldt_sel;#ifdef CONFIG_X86_64 u16 ds_sel, es_sel;
@@ -2750,10 +2754,11 @@ static void vmx_save_host_state(struct k#endif int i; -if (vmx->host_state.loaded) +if (vmx->loaded_cpu_state)return ; - vmx->host_state.loaded =1 ; + vmx->loaded_cpu_state = vmx->loaded_vmcs; +/* * Set host fs and gs selectors. Unfortunately, 22.2.3 does not * allow segment selectors with cpl > 0 or ti == 1.
@@ -2815,11 +2820,14 @@ static void vmx_save_host_state(struct kstatic void __vmx_load_host_state (struct vcpu_vmx *vmx) { -if (!vmx->host_state.loaded) +if (!vmx->loaded_cpu_state)return ; +WARN_ON_ONCE (vmx->loaded_cpu_state != vmx->loaded_vmcs); + ++vmx->vcpu.stat.host_state_reload; - vmx->host_state.loaded =0 ; + vmx->loaded_cpu_state =NULL ; +#ifdef CONFIG_X86_64 if (is_long_mode (&vmx->vcpu))rdmsrl (MSR_KERNEL_GS_BASE , vmx->msr_guest_kernel_gs_base);
@@ -10517,8 +10525,8 @@ static void vmx_switch_vmcs(struct kvm_vreturn ; cpu =get_cpu (); - vmx->loaded_vmcs = vmcs;vmx_vcpu_put (vcpu); + vmx->loaded_vmcs = vmcs;vmx_vcpu_load (vcpu, cpu);put_cpu (); }
Greg Kroah-Hartman
maintainerPATCH 131/197 kvm: nVMX: Fix fault vector for VMX operation at CPL > 0
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jim Mattson <[email protected]>
[ Upstream commit 36090bf43a6b835a42f515cb515ff6fa293a25fe ]
The fault that should be raised for a privilege level violation is #GP rather than #UD.
--- arch/x86/kvm/vmx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/vmx.c +2 -2
@@ -8123,7 +8123,7 @@ static int handle_vmon(struct kvm_vcpu */* CPL=0 must be checked manually. */ if (vmx_get_cpl (vcpu)) { -kvm_queue_exception (vcpu,UD_VECTOR ); +kvm_inject_gp (vcpu,0 );return 1 ; }
@@ -8187,7 +8187,7 @@ static int handle_vmon(struct kvm_vcpu *static int nested_vmx_check_permission (struct kvm_vcpu *vcpu) {if (vmx_get_cpl (vcpu)) { -kvm_queue_exception (vcpu,UD_VECTOR ); +kvm_inject_gp (vcpu,0 );return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 132/197 drm/etnaviv: fix crash in GPU suspend when init failed due to buffer placement
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lucas Stach <[email protected]>
[ Upstream commit 5b147465532365dc4e2fee8499d6ca1f52dd0d16 ]
When the suballocator was unable to provide a suitable buffer for the MMUv1 linear window, we roll back the GPU initialization. As the GPU is runtime resumed at that point we need to clear the kernel cmdbuf suballoc entry to properly skip any attempt to manipulate the cmdbuf when the GPU gets shut down in the runtime suspend later on.
--- drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/etnaviv/etnaviv_gpu.c +1 -0
@@ -799,6 +799,7 @@ int etnaviv_gpu_init(struct etnaviv_gpu free_buffer:etnaviv_cmdbuf_free (&gpu->buffer); + gpu->buffer.suballoc =NULL ; destroy_iommu:etnaviv_iommu_destroy (gpu->mmu); gpu->mmu =NULL ;
Greg Kroah-Hartman
maintainerPATCH 133/197 btrfs: Exit gracefully when chunk map cannot be inserted to the tree
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Qu Wenruo <[email protected]>
[ Upstream commit 64f64f43c89aca1782aa672e0586f6903c5d8979 ]
It's entirely possible that a crafted btrfs image contains overlapping chunks.
Although we can't detect such problem by tree-checker, it's not a catastrophic problem, current extent map can already detect such problem and return -EEXIST.
We just only need to exit gracefully and fail the mount.
--- fs/btrfs/volumes.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
fs/btrfs/volumes.c +6 -2
@@ -6563,10 +6563,14 @@ static int read_one_chunk(struct btrfs_fwrite_lock (&map_tree->map_tree.lock); ret =add_extent_mapping (&map_tree->map_tree, em,0 );write_unlock (&map_tree->map_tree.lock); -BUG_ON (ret);/* Tree corruption */ +if (ret <0 ) { +btrfs_err (fs_info, +"failed to add chunk map, start=%llu len=%llu: %d" , + em->start, em->len, ret); + }free_extent_map (em); -return 0 ; +return ret; } static void fill_device_from_item(struct extent_buffer *leaf,
Greg Kroah-Hartman
maintainerPATCH 134/197 btrfs: replace: Reset on-disk dev stats value after replace
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Misono Tomohiro <[email protected]>
[ Upstream commit 1e7e1f9e3aba00c9b9c323bfeeddafe69ff21ff6 ]
on-disk devs stats value is updated in btrfs_run_dev_stats(), which is called during commit transaction, if device->dev_stats_ccnt is not zero.
Since current replace operation does not touch dev_stats_ccnt, on-disk dev stats value is not updated. Therefore "btrfs device stats" may return old device's value after umount/mount
(Example: See "btrfs ins dump-t -t DEV $DEV" after btrfs/100 finish).
Fix this by just incrementing dev_stats_ccnt in btrfs_dev_replace_finishing() when replace is succeeded and this will update the values.
--- fs/btrfs/dev-replace.c | 6 ++++++ 1 file changed, 6 insertions(+)
fs/btrfs/dev-replace.c +6 -0
@@ -677,6 +677,12 @@ static int btrfs_dev_replace_finishing(sbtrfs_rm_dev_replace_unblocked (fs_info);/* +* Increment dev_stats_ccnt so that btrfs_run_dev_stats() will +* update on-disk dev stats value during commit transaction +*/ +atomic_inc (&tgt_device->dev_stats_ccnt); + +/* * this is again a consistent state where no dev_replace procedure * is running, the target device is part of the filesystem, the * source device is not part of the filesystem anymore and its 1st
Greg Kroah-Hartman
maintainerPATCH 135/197 btrfs: fix in-memory value of total_devices after seed device deletion
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anand Jain <[email protected]>
[ Upstream commit b4993e64f78a9605b45252fa9ba385c88a1f4ce9 ]
In case of deleting the seed device the %cur_devices (seed) and the %fs_devices (parent) are different. Now, as the parent fs_devices::total_devices also maintains the total number of devices including the seed device, so decrement its in-memory value for the successful seed delete. We are already updating its corresponding on-disk btrfs_super_block::number_devices value.
--- fs/btrfs/volumes.c | 3 +++ 1 file changed, 3 insertions(+)
fs/btrfs/volumes.c +3 -0
@@ -2029,6 +2029,9 @@ int btrfs_rm_device(struct btrfs_fs_info cur_devices->num_devices--; cur_devices->total_devices--; +/* Update total_devices of the parent fs_devices if it's seed */ +if (cur_devices != fs_devices) + fs_devices->total_devices--;if (test_bit (BTRFS_DEV_STATE_MISSING , &device->dev_state)) cur_devices->missing_devices--;
Greg Kroah-Hartman
maintainerPATCH 136/197 btrfs: relocation: Only remove reloc rb_trees if reloc control has been initialized
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Qu Wenruo <[email protected]>
[ Upstream commit 389305b2aa68723c754f88d9dbd268a400e10664 ]
Invalid reloc tree can cause kernel NULL pointer dereference when btrfs does some cleanup of the reloc roots.
It turns out that fs_info::reloc_ctl can be NULL in btrfs_recover_relocation() as we allocate relocation control after all reloc roots have been verified.
So when we hit: note, we haven't called set_reloc_control() thus fs_info::reloc_ctl is still NULL.
--- fs/btrfs/relocation.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-)
fs/btrfs/relocation.c +12 -11
@@ -1321,18 +1321,19 @@ static void __del_reloc_root(struct btrfstruct mapping_node *node =NULL ;struct reloc_control *rc = fs_info->reloc_ctl; -spin_lock (&rc->reloc_root_tree.lock); - rb_node =tree_search (&rc->reloc_root_tree.rb_root, - root->node->start); -if (rb_node) { - node = rb_entry(rb_node, struct mapping_node, rb_node); -rb_erase (&node->rb_node, &rc->reloc_root_tree.rb_root); +if (rc) { +spin_lock (&rc->reloc_root_tree.lock); + rb_node =tree_search (&rc->reloc_root_tree.rb_root, + root->node->start); +if (rb_node) { + node = rb_entry(rb_node, struct mapping_node, rb_node); +rb_erase (&node->rb_node, &rc->reloc_root_tree.rb_root); + } +spin_unlock (&rc->reloc_root_tree.lock); +if (!node) +return ; +BUG_ON ((struct btrfs_root *)node->data != root); } -spin_unlock (&rc->reloc_root_tree.lock); - -if (!node) -return ; -BUG_ON ((struct btrfs_root *)node->data != root);spin_lock (&fs_info->trans_lock);list_del_init (&root->root_list);
Greg Kroah-Hartman
maintainerPATCH 137/197 btrfs: tree-checker: Detect invalid and empty essential trees
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Qu Wenruo <[email protected]>
[ Upstream commit ba480dd4db9f1798541eb2d1c423fc95feee8d36 ]
A crafted image has empty root tree block, which will later cause NULL pointer dereference.
The following trees should never be empty:
1) Tree root
Must contain at least root items for extent tree, device tree and fs tree
2) Chunk tree
Or we can't even bootstrap as it contains the mapping.
3) Fs tree
At least inode item for top level inode (.).
4) Device tree
Dev extents for chunks
5) Extent tree
Must have corresponding extent for each chunk.
If any of them is empty, we are sure the fs is corrupted and no need to mount it.
--- fs/btrfs/tree-checker.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
fs/btrfs/tree-checker.c +14 -1
@@ -396,9 +396,22 @@ static int check_leaf(struct btrfs_fs_in* skip this check for relocation trees. */ if (nritems ==0 && !btrfs_header_flag (leaf,BTRFS_HEADER_FLAG_RELOC )) { +u64 owner =btrfs_header_owner (leaf);struct btrfs_root *check_root; - key.objectid =btrfs_header_owner (leaf); +/* These trees must never be empty */ +if (owner ==BTRFS_ROOT_TREE_OBJECTID || + owner ==BTRFS_CHUNK_TREE_OBJECTID || + owner ==BTRFS_EXTENT_TREE_OBJECTID || + owner ==BTRFS_DEV_TREE_OBJECTID || + owner ==BTRFS_FS_TREE_OBJECTID || + owner ==BTRFS_DATA_RELOC_TREE_OBJECTID ) { +generic_err (fs_info, leaf,0 , +"invalid root, root %llu must never be empty" , + owner); +return -EUCLEAN ; + } + key.objectid = owner; key.type =BTRFS_ROOT_ITEM_KEY ; key.offset = (u64)-1 ;
Greg Kroah-Hartman
maintainerPATCH 138/197 btrfs: check-integrity: Fix NULL pointer dereference for degraded mount
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Qu Wenruo <[email protected]>
[ Upstream commit 9912bbf6440ba0555e91d3306520da01872c7c1d ]
Commit f8f84b2dfda5 ("btrfs: index check-integrity state hash by a dev_t") changed how btrfsic indexes device state.
Now we need to access device->bdev->bd_dev, while for degraded mount it's completely possible to have device->bdev as NULL, thus it will trigger a NULL pointer dereference at mount time.
Fix it by checking if the device is degraded before accessing device->bdev->bd_dev.
There are a lot of other places accessing device->bdev->bd_dev, however the other call sites have either checked device->bdev, or the device->bdev is passed from btrfsic_map_block(), so it won't cause harm.
--- fs/btrfs/check-integrity.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
fs/btrfs/check-integrity.c +6 -1
@@ -1539,7 +1539,12 @@ static int btrfsic_map_block(struct btrf } device = multi->stripes[0 ].dev; - block_ctx_out->dev =btrfsic_dev_state_lookup (device->bdev->bd_dev); +if (test_bit (BTRFS_DEV_STATE_MISSING , &device->dev_state) || + !device->bdev || !device->name) + block_ctx_out->dev =NULL ; +else + block_ctx_out->dev =btrfsic_dev_state_lookup ( + device->bdev->bd_dev); block_ctx_out->dev_bytenr = multi->stripes[0 ].physical; block_ctx_out->start = bytenr; block_ctx_out->len = len;
Greg Kroah-Hartman
maintainerPATCH 139/197 btrfs: lift uuid_mutex to callers of btrfs_open_devices
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Sterba <[email protected]>
[ Upstream commit f5194e34cabaddd348a90f950e0a8188dd26cdc0 ]
Prepartory work to fix race between mount and device scan.
The callers will have to manage the critical section, eg. mount wants to scan and then call btrfs_open_devices without the ioctl scan walking in and modifying the fs devices in the meantime.
--- fs/btrfs/super.c | 2 ++ fs/btrfs/volumes.c | 4 ++-- 2 files changed, 4 insertions(+), 2 deletions(-)
fs/btrfs/super.c +2 -0
@@ -1565,7 +1565,9 @@ static struct dentry *btrfs_mount_root(s goto error_fs_info; } +mutex_lock (&uuid_mutex); error =btrfs_open_devices (fs_devices, mode, fs_type); +mutex_unlock (&uuid_mutex);if (error) goto error_fs_info;
fs/btrfs/volumes.c +2 -2
@@ -1146,7 +1146,8 @@ int btrfs_open_devices(struct btrfs_fs_d {int ret; -mutex_lock (&uuid_mutex); +lockdep_assert_held (&uuid_mutex); +mutex_lock (&fs_devices->device_list_mutex);if (fs_devices->opened) { fs_devices->opened++;
@@ -1156,7 +1157,6 @@ int btrfs_open_devices(struct btrfs_fs_d ret =open_fs_devices (fs_devices, flags, holder); }mutex_unlock (&fs_devices->device_list_mutex); -mutex_unlock (&uuid_mutex);return ret; }
Greg Kroah-Hartman
maintainerPATCH 140/197 btrfs: Dont remove block group that still has pinned down bytes
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Qu Wenruo <[email protected]>
[ Upstream commit 43794446548730ac8461be30bbe47d5d027d1d16 ]
[BUG]
Under certain KVM load and LTP tests, it is possible to hit the following calltrace if quota is enabled:
BTRFS critical (device vda2): unable to find logical 8820195328 length 4096 BTRFS critical (device vda2): unable to find logical 8820195328 length 4096
WARNING: CPU: 0 PID: 49 at ../block/blk-core.c:172 blk_status_to_errno+0x1a/0x30 CPU: 0 PID: 49 Comm: kworker/u2:1 Not tainted 4.12.14-15-default #1 SLE15 (unreleased) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org 04/01/2014 Workqueue: btrfs-endio-write btrfs_endio_write_helper [btrfs] task: ffff9f827b340bc0 task.stack: ffffb4f8c0304000 RIP: 0010:blk_status_to_errno+0x1a/0x30 Call Trace: submit_extent_page+0x191/0x270 [btrfs] ? btrfs_create_repair_bio+0x130/0x130 [btrfs] __do_readpage+0x2d2/0x810 [btrfs] ? btrfs_create_repair_bio+0x130/0x130 [btrfs] ? run_one_async_done+0xc0/0xc0 [btrfs] __extent_read_full_page+0xe7/0x100 [btrfs] ? run_one_async_done+0xc0/0xc0 [btrfs] read_extent_buffer_pages+0x1ab/0x2d0 [btrfs] ? run_one_async_done+0xc0/0xc0 [btrfs] btree_read_extent_buffer_pages+0x94/0xf0 [btrfs] read_tree_block+0x31/0x60 [btrfs] read_block_for_search.isra.35+0xf0/0x2e0 [btrfs] btrfs_search_slot+0x46b/0xa00 [btrfs] ? kmem_cache_alloc+0x1a8/0x510 ? btrfs_get_token_32+0x5b/0x120 [btrfs] find_parent_nodes+0x11d/0xeb0 [btrfs] ? leaf_space_used+0xb8/0xd0 [btrfs] ? btrfs_leaf_free_space+0x49/0x90 [btrfs] ? btrfs_find_all_roots_safe+0x93/0x100 [btrfs] btrfs_find_all_roots_safe+0x93/0x100 [btrfs] btrfs_find_all_roots+0x45/0x60 [btrfs] btrfs_qgroup_trace_extent_post+0x20/0x40 [btrfs] btrfs_add_delayed_data_ref+0x1a3/0x1d0 [btrfs] btrfs_alloc_reserved_file_extent+0x38/0x40 [btrfs] insert_reserved_file_extent.constprop.71+0x289/0x2e0 [btrfs] btrfs_finish_ordered_io+0x2f4/0x7f0 [btrfs] ? pick_next_task_fair+0x2cd/0x530 ? __switch_to+0x92/0x4b0 btrfs_worker_helper+0x81/0x300 [btrfs] process_one_work+0x1da/0x3f0 worker_thread+0x2b/0x3f0 ? process_one_work+0x3f0/0x3f0 kthread+0x11a/0x130 ? kthread_create_on_node+0x40/0x40 ret_from_fork+0x35/0x40
BTRFS critical (device vda2): unable to find logical 8820195328 length 16384
BTRFS: error (device vda2) in btrfs_finish_ordered_io:3023: errno=-5 IO failure BTRFS info (device vda2): forced readonly
BTRFS error (device vda2): pending csums is 2887680
[CAUSE]
It's caused by race with block group auto removal:
- There is a meta block group X, which has only one tree block The tree block belongs to fs tree 257.
- In current transaction, some operation modified fs tree 257 The tree block gets COWed, so the block group X is empty, and marked as unused, queued to be deleted.
- Some workload (like fsync) wakes up cleaner_kthread() Which will call btrfs_delete_unused_bgs() to remove unused block groups.
So block group X along its chunk map get removed.
- Some delalloc work finished for fs tree 257
Quota needs to get the original reference of the extent, which will read tree blocks of commit root of 257.
Then since the chunk map gets removed, the above warning gets triggered.
[FIX]
Just let btrfs_delete_unused_bgs() skip block group which still has pinned bytes.
However there is a minor side effect: currently we only queue empty blocks at update_block_group(), and such empty block group with pinned bytes won't go through update_block_group() again, such block group won't be removed, until it gets new extent allocated and removed.
--- fs/btrfs/extent-tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/extent-tree.c +1 -1
@@ -10687,7 +10687,7 @@ void btrfs_delete_unused_bgs(struct btrf/* Don't want to race with allocators so take the groups_sem */ down_write (&space_info->groups_sem);spin_lock (&block_group->lock); -if (block_group->reserved || +if (block_group->reserved || block_group->pinned ||btrfs_block_group_used (&block_group->item) || block_group->ro ||list_is_singular (&block_group->list)) {
Greg Kroah-Hartman
maintainerPATCH 141/197 btrfs: Fix a C compliance issue
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Bart Van Assche <[email protected]>
[ Upstream commit edf57cbf2b030781885e339f32e35a470d2f8eba ]
The C programming language does not allow to use preprocessor statements inside macro arguments (pr_info() is defined as a macro). Hence rework the pr_info() statement in btrfs_print_mod_info() such that it becomes compliant. This patch allows tools like sparse to analyze the BTRFS source code.
--- fs/btrfs/super.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
fs/btrfs/super.c +3 -3
@@ -2370,7 +2370,7 @@ static __cold void btrfs_interface_exit(static void __init btrfs_print_mod_info (void ) { - pr_info("Btrfs loaded, crc32c=%s" +static const char options[] ="" #ifdef CONFIG_BTRFS_DEBUG", debug=on" #endif
@@ -2383,8 +2383,8 @@ static void __init btrfs_print_mod_info(#ifdef CONFIG_BTRFS_FS_REF_VERIFY ", ref-verify=on"#endif -"\n" , - crc32c_impl()); + ; +pr_info ("Btrfs loaded, crc32c=%s%s\n" ,crc32c_impl (), options); }static int __init init_btrfs_fs (void )
Greg Kroah-Hartman
maintainerPATCH 142/197 arm64: rockchip: Force CONFIG_PM on Rockchip systems
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marc Zyngier <[email protected]>
[ Upstream commit 7db7a8f5638a2ffe0c0c0d55b5186b6191fd6af7 ]
A number of the Rockchip-specific drivers (IOMMU, display controllers) are now assuming that CONFIG_PM is set, and may completely misbehave if that's not the case.
Since there is hardly any reason for this configuration option not to be selected anyway, let's require it (in the same way Tegra already does).
--- arch/arm64/Kconfig.platforms | 1 + 1 file changed, 1 insertion(+)
arch/arm64/Kconfig.platforms +1 -0
@@ -151,6 +151,7 @@ config ARCH_ROCKCHIP select GPIOLIB select PINCTRL select PINCTRL_ROCKCHIP + select PM select ROCKCHIP_TIMER help This enables support for the ARMv8 based Rockchip chipsets,
Greg Kroah-Hartman
maintainerPATCH 143/197 ARM: rockchip: Force CONFIG_PM on Rockchip systems
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marc Zyngier <[email protected]>
[ Upstream commit d1558dfd9f22c99a5b8e1354ad881ee40749da89 ]
A number of the Rockchip-specific drivers (IOMMU, display controllers) are now assuming that CONFIG_PM is set, and may completely misbehave if that's not the case.
Since there is hardly any reason for this configuration option not to be selected anyway, let's require it (in the same way Tegra already does).
--- arch/arm/mach-rockchip/Kconfig | 1 + 1 file changed, 1 insertion(+)
arch/arm/mach-rockchip/Kconfig +1 -0
@@ -17,6 +17,7 @@ config ARCH_ROCKCHIP select ARM_GLOBAL_TIMER select CLKSRC_ARM_GLOBAL_TIMER_SCHED_CLOCK select ZONE_DMA if ARM_LPAE + select PM help Support for Rockchip's Cortex-A9 Single-to-Quad-Core-SoCs containing the RK2928, RK30xx and RK31xx series.
Greg Kroah-Hartman
maintainerPATCH 144/197 btrfs: do btrfs_free_stale_devices outside of device_list_add
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anand Jain <[email protected]>
[ Upstream commit 4306a97449f9a0f9e5229af7889d4401315355aa ]
btrfs_free_stale_devices() looks for device path reused for another filesystem, and deletes the older fs_devices::device entry.
In preparation to handle locking in device_list_add, move btrfs_free_stale_devices outside as these two functions serve a different purpose.
--- fs/btrfs/volumes.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-)
fs/btrfs/volumes.c +10 -5
@@ -750,7 +750,8 @@ error_brelse:* error pointer when failed */ static noinline struct btrfs_device *device_list_add(const char *path, -struct btrfs_super_block *disk_super) +struct btrfs_super_block *disk_super, +bool *new_device_added) {struct btrfs_device *device;struct btrfs_fs_devices *fs_devices;
@@ -796,7 +797,7 @@ static noinline struct btrfs_device *devmutex_unlock (&fs_devices->device_list_mutex); device->fs_devices = fs_devices; -btrfs_free_stale_devices (path, device); + *new_device_added = true;if (disk_super->label[0 ]) pr_info("BTRFS: device label %s devid %llu transid %llu %s\n",
@@ -1221,6 +1222,7 @@ int btrfs_scan_one_device(const char *pa struct btrfs_fs_devices **fs_devices_ret) {struct btrfs_super_block *disk_super; +bool new_device_added = false;struct btrfs_device *device;struct block_device *bdev;struct page *page;
@@ -1246,11 +1248,14 @@ int btrfs_scan_one_device(const char *pa }mutex_lock (&uuid_mutex); - device =device_list_add (path, disk_super); -if (IS_ERR (device)) + device =device_list_add (path, disk_super, &new_device_added); +if (IS_ERR (device)) { ret =PTR_ERR (device); -else + }else { *fs_devices_ret = device->fs_devices; +if (new_device_added) +btrfs_free_stale_devices (path, device); + }mutex_unlock (&uuid_mutex);btrfs_release_disk_super (page);
Greg Kroah-Hartman
maintainerPATCH 145/197 btrfs: extend locked section when adding a new device in device_list_add
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anand Jain <[email protected]>
[ Upstream commit 9c6d173ea6e4c8c939ae6c257c7fc18f7b320316 ]
Make sure the device_list_lock is held the whole time:
* when the device is being looked up
* new device is initialized and put to the list
* the list counters are updated (fs_devices::opened, fs_devices::total_devices)
--- fs/btrfs/volumes.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)
fs/btrfs/volumes.c +12 -4
@@ -765,21 +765,26 @@ static noinline struct btrfs_device *devif (IS_ERR (fs_devices))return ERR_CAST (fs_devices); +mutex_lock (&fs_devices->device_list_mutex);list_add (&fs_devices->fs_list, &fs_uuids); device =NULL ; }else { +mutex_lock (&fs_devices->device_list_mutex); device =find_device (fs_devices, devid, disk_super->dev_item.uuid); }if (!device) { -if (fs_devices->opened) +if (fs_devices->opened) { +mutex_unlock (&fs_devices->device_list_mutex);return ERR_PTR (-EBUSY ); + } device =btrfs_alloc_device (NULL , &devid, disk_super->dev_item.uuid);if (IS_ERR (device)) { +mutex_unlock (&fs_devices->device_list_mutex);/* we can safely leave the fs_devices entry around */ return device; }
@@ -787,14 +792,13 @@ static noinline struct btrfs_device *dev name =rcu_string_strdup (path,GFP_NOFS );if (!name) {btrfs_free_device (device); +mutex_unlock (&fs_devices->device_list_mutex);return ERR_PTR (-ENOMEM ); }rcu_assign_pointer (device->name, name); -mutex_lock (&fs_devices->device_list_mutex);list_add_rcu (&device->dev_list, &fs_devices->devices); fs_devices->num_devices++; -mutex_unlock (&fs_devices->device_list_mutex); device->fs_devices = fs_devices; *new_device_added = true;
@@ -841,12 +845,15 @@ static noinline struct btrfs_device *dev* with larger generation number or the last-in if * generation are equal. */ +mutex_unlock (&fs_devices->device_list_mutex);return ERR_PTR (-EEXIST ); } name =rcu_string_strdup (path,GFP_NOFS ); -if (!name) +if (!name) { +mutex_unlock (&fs_devices->device_list_mutex);return ERR_PTR (-ENOMEM ); + }rcu_string_free (device->name);rcu_assign_pointer (device->name, name);if (test_bit (BTRFS_DEV_STATE_MISSING , &device->dev_state)) {
@@ -866,6 +873,7 @@ static noinline struct btrfs_device *dev fs_devices->total_devices =btrfs_super_num_devices (disk_super); +mutex_unlock (&fs_devices->device_list_mutex);return device; }
Greg Kroah-Hartman
maintainerPATCH 146/197 btrfs: rename local devices for fs_devices in btrfs_free_stale_devices(
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anand Jain <[email protected]>
[ Upstream commit fa6d2ae540a200a17bb7ee769f9df22d411c9404 ]
Over the years we named %fs_devices and %devices to represent the struct btrfs_fs_devices and the struct btrfs_device. So follow the same scheme here too. No functional changes.
--- fs/btrfs/volumes.c | 35 +++++++++++++++++------------------ 1 file changed, 17 insertions(+), 18 deletions(-)
fs/btrfs/volumes.c +17 -18
@@ -634,43 +634,42 @@ static void pending_bios_fn(struct btrfs* devices. */ static void btrfs_free_stale_devices (const char *path, -struct btrfs_device *skip_dev) +struct btrfs_device *skip_device) { -struct btrfs_fs_devices *fs_devs, *tmp_fs_devs; -struct btrfs_device *dev, *tmp_dev; +struct btrfs_fs_devices *fs_devices, *tmp_fs_devices; +struct btrfs_device *device, *tmp_device; -list_for_each_entry_safe (fs_devs, tmp_fs_devs, &fs_uuids, fs_list) { - -if (fs_devs->opened) +list_for_each_entry_safe (fs_devices, tmp_fs_devices, &fs_uuids, fs_list) { +if (fs_devices->opened)continue ; -list_for_each_entry_safe (dev, tmp_dev, - &fs_devs->devices, dev_list) { +list_for_each_entry_safe (device, tmp_device, + &fs_devices->devices, dev_list) {int not_found =0 ; -if (skip_dev && skip_dev == dev) +if (skip_device && skip_device == device)continue ; -if (path && !dev->name) +if (path && !device->name)continue ;rcu_read_lock ();if (path) - not_found =strcmp (rcu_str_deref (dev->name), + not_found =strcmp (rcu_str_deref (device->name), path);rcu_read_unlock ();if (not_found)continue ;/* delete the stale device */ -if (fs_devs->num_devices ==1 ) { -btrfs_sysfs_remove_fsid (fs_devs); -list_del (&fs_devs->fs_list); -free_fs_devices (fs_devs); +if (fs_devices->num_devices ==1 ) { +btrfs_sysfs_remove_fsid (fs_devices); +list_del (&fs_devices->fs_list); +free_fs_devices (fs_devices);break ; }else { - fs_devs->num_devices--; -list_del (&dev->dev_list); -btrfs_free_device (dev); + fs_devices->num_devices--; +list_del (&device->dev_list); +btrfs_free_device (device); } } }
Greg Kroah-Hartman
maintainerPATCH 147/197 btrfs: use device_list_mutex when removing stale devices
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Anand Jain <[email protected]>
[ Upstream commit 7bcb8164ad9435068d9bc3b83b8a002c64d63ff6 ]
btrfs_free_stale_devices() finds a stale (not opened) device matching path in the fs_uuid list. We are already under uuid_mutex so when we check for each fs_devices, hold the device_list_mutex too.
--- fs/btrfs/volumes.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-)
fs/btrfs/volumes.c +15 -10
@@ -640,8 +640,11 @@ static void btrfs_free_stale_devices(construct btrfs_device *device, *tmp_device;list_for_each_entry_safe (fs_devices, tmp_fs_devices, &fs_uuids, fs_list) { -if (fs_devices->opened) +mutex_lock (&fs_devices->device_list_mutex); +if (fs_devices->opened) { +mutex_unlock (&fs_devices->device_list_mutex);continue ; + }list_for_each_entry_safe (device, tmp_device, &fs_devices->devices, dev_list) {
@@ -661,16 +664,18 @@ static void btrfs_free_stale_devices(concontinue ;/* delete the stale device */ -if (fs_devices->num_devices ==1 ) { -btrfs_sysfs_remove_fsid (fs_devices); -list_del (&fs_devices->fs_list); -free_fs_devices (fs_devices); + fs_devices->num_devices--; +list_del (&device->dev_list); +btrfs_free_device (device); + +if (fs_devices->num_devices ==0 )break ; - }else { - fs_devices->num_devices--; -list_del (&device->dev_list); -btrfs_free_device (device); - } + } +mutex_unlock (&fs_devices->device_list_mutex); +if (fs_devices->num_devices ==0 ) { +btrfs_sysfs_remove_fsid (fs_devices); +list_del (&fs_devices->fs_list); +free_fs_devices (fs_devices); } } }
Greg Kroah-Hartman
maintainerPATCH 148/197 btrfs: lift uuid_mutex to callers of btrfs_scan_one_device
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Sterba <[email protected]>
[ Upstream commit 899f9307c33ce4758c30a076b10ed54d5c91c6e7 ]
Prepartory work to fix race between mount and device scan.
The callers will have to manage the critical section, eg. mount wants to scan and then call btrfs_open_devices without the ioctl scan walking in and modifying the fs devices in the meantime.
--- fs/btrfs/super.c | 12 +++++++++++- fs/btrfs/volumes.c | 4 ++-- 2 files changed, 13 insertions(+), 3 deletions(-)
fs/btrfs/super.c +11 -1
@@ -917,8 +917,10 @@ static int btrfs_parse_early_options(con error = -ENOMEM ; goto out; } +mutex_lock (&uuid_mutex); error =btrfs_scan_one_device (device_name, flags, holder, fs_devices); +mutex_unlock (&uuid_mutex);kfree (device_name);if (error) goto out;
@@ -1539,7 +1541,9 @@ static struct dentry *btrfs_mount_root(sreturn ERR_PTR (error); } +mutex_lock (&uuid_mutex); error =btrfs_scan_one_device (device_name, mode, fs_type, &fs_devices); +mutex_unlock (&uuid_mutex);if (error) goto error_sec_opts;
@@ -2236,15 +2240,21 @@ static long btrfs_control_ioctl(struct fswitch (cmd) {case BTRFS_IOC_SCAN_DEV : +mutex_lock (&uuid_mutex); ret =btrfs_scan_one_device (vol->name,FMODE_READ , &btrfs_root_fs_type, &fs_devices); +mutex_unlock (&uuid_mutex);break ;case BTRFS_IOC_DEVICES_READY : +mutex_lock (&uuid_mutex); ret =btrfs_scan_one_device (vol->name,FMODE_READ , &btrfs_root_fs_type, &fs_devices); -if (ret) +if (ret) { +mutex_unlock (&uuid_mutex);break ; + } ret = !(fs_devices->num_devices == fs_devices->total_devices); +mutex_unlock (&uuid_mutex);break ;case BTRFS_IOC_GET_SUPPORTED_FEATURES : ret =btrfs_ioctl_get_supported_features ((void __user *)arg);
fs/btrfs/volumes.c +2 -2
@@ -1241,6 +1241,8 @@ int btrfs_scan_one_device(const char *paint ret =0 ;u64 bytenr; +lockdep_assert_held (&uuid_mutex); +/* * we would like to check all the supers, but that would make * a btrfs mount succeed after a mkfs from a different FS.
@@ -1259,7 +1261,6 @@ int btrfs_scan_one_device(const char *pa goto error_bdev_put; } -mutex_lock (&uuid_mutex); device =device_list_add (path, disk_super, &new_device_added);if (IS_ERR (device)) { ret =PTR_ERR (device);
@@ -1268,7 +1269,6 @@ int btrfs_scan_one_device(const char *paif (new_device_added)btrfs_free_stale_devices (path, device); } -mutex_unlock (&uuid_mutex);btrfs_release_disk_super (page);
Greg Kroah-Hartman
maintainerPATCH 149/197 btrfs: lift uuid_mutex to callers of btrfs_parse_early_options
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Sterba <[email protected]>
[ Upstream commit 5139cff598d42b1e531f40c84691a7e945f04553 ]
Prepartory work to fix race between mount and device scan.
btrfs_parse_early_options calls the device scan from mount and we'll need to let mount completely manage the critical section.
--- fs/btrfs/super.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
fs/btrfs/super.c +4 -2
@@ -892,6 +892,8 @@ static int btrfs_parse_early_options(conchar *device_name, *opts, *orig, *p;int error =0 ; +lockdep_assert_held (&uuid_mutex); +if (!options)return 0 ;
@@ -917,10 +919,8 @@ static int btrfs_parse_early_options(con error = -ENOMEM ; goto out; } -mutex_lock (&uuid_mutex); error =btrfs_scan_one_device (device_name, flags, holder, fs_devices); -mutex_unlock (&uuid_mutex);kfree (device_name);if (error) goto out;
@@ -1528,8 +1528,10 @@ static struct dentry *btrfs_mount_root(sif (!(flags &SB_RDONLY )) mode |=FMODE_WRITE ; +mutex_lock (&uuid_mutex); error =btrfs_parse_early_options (data, mode, fs_type, &fs_devices); +mutex_unlock (&uuid_mutex);if (error) {return ERR_PTR (error); }
Greg Kroah-Hartman
maintainerPATCH 150/197 btrfs: reorder initialization before the mount locks uuid_mutex
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Sterba <[email protected]>
[ Upstream commit 399f7f4c42e8a58c8456264d5112287aefe44cf4 ]
In preparation to take a big lock, move resource initialization before the critical section. It's not obvious from the diff, the desired order is:
- initialize mount security options
- allocate temporary fs_info
- allocate superblock buffers
--- fs/btrfs/super.c | 30 ++++++++++++++---------------- 1 file changed, 14 insertions(+), 16 deletions(-)
fs/btrfs/super.c +14 -16
@@ -1528,14 +1528,6 @@ static struct dentry *btrfs_mount_root(sif (!(flags &SB_RDONLY )) mode |=FMODE_WRITE ; -mutex_lock (&uuid_mutex); - error =btrfs_parse_early_options (data, mode, fs_type, - &fs_devices); -mutex_unlock (&uuid_mutex); -if (error) { -return ERR_PTR (error); - } -security_init_mnt_opts (&new_sec_opts);if (data) { error =parse_security_options (data, &new_sec_opts);
@@ -1543,12 +1535,6 @@ static struct dentry *btrfs_mount_root(sreturn ERR_PTR (error); } -mutex_lock (&uuid_mutex); - error =btrfs_scan_one_device (device_name, mode, fs_type, &fs_devices); -mutex_unlock (&uuid_mutex); -if (error) - goto error_sec_opts; -/* * Setup a dummy root and fs_info for test/set super. This is because * we don't actually fill this stuff out until open_ctree, but we need
@@ -1561,8 +1547,6 @@ static struct dentry *btrfs_mount_root(s goto error_sec_opts; } - fs_info->fs_devices = fs_devices; - fs_info->super_copy =kzalloc (BTRFS_SUPER_INFO_SIZE ,GFP_KERNEL ); fs_info->super_for_commit =kzalloc (BTRFS_SUPER_INFO_SIZE ,GFP_KERNEL );security_init_mnt_opts (&fs_info->security_opts);
@@ -1572,6 +1556,20 @@ static struct dentry *btrfs_mount_root(s }mutex_lock (&uuid_mutex); + error =btrfs_parse_early_options (data, mode, fs_type, &fs_devices); +mutex_unlock (&uuid_mutex); +if (error) + goto error_fs_info; + +mutex_lock (&uuid_mutex); + error =btrfs_scan_one_device (device_name, mode, fs_type, &fs_devices); +mutex_unlock (&uuid_mutex); +if (error) + goto error_fs_info; + + fs_info->fs_devices = fs_devices; + +mutex_lock (&uuid_mutex); error =btrfs_open_devices (fs_devices, mode, fs_type);mutex_unlock (&uuid_mutex);if (error)
Greg Kroah-Hartman
maintainerPATCH 151/197 btrfs: fix mount and ioctl device scan ioctl race
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Sterba <[email protected]>
[ Upstream commit 81ffd56b5745355b70d54ca4e1bdd0d64a66ff9f ]
Technically this extends the critical section covered by uuid_mutex to:
- parse early mount options -- here we can call device scan on paths that can be passed as 'device=/dev/...'
- scan the device passed to mount
- open the devices related to the fs_devices -- this increases fs_devices::opened
The race can happen when mount calls one of the scans and there's another one called eg. by mkfs or 'btrfs dev scan':
Mount Scan
----- ----scan_one_device (dev1, fsid1)
scan_one_device (dev2, fsid1)
add the device
free stale devices
fsid1 fs_devices::opened == 0
find fsid1:dev1 free fsid1:dev1 if it's the last one,
free fs_devices of fsid1 too
open_devices (dev1, fsid1)
dev1 not found
When fixed, the uuid mutex will make sure that mount will increase fs_devices::opened and this will not be touched by the racing scan ioctl.
--- fs/btrfs/super.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
fs/btrfs/super.c +6 -6
@@ -1557,19 +1557,19 @@ static struct dentry *btrfs_mount_root(smutex_lock (&uuid_mutex); error =btrfs_parse_early_options (data, mode, fs_type, &fs_devices); -mutex_unlock (&uuid_mutex); -if (error) +if (error) { +mutex_unlock (&uuid_mutex); goto error_fs_info; + } -mutex_lock (&uuid_mutex); error =btrfs_scan_one_device (device_name, mode, fs_type, &fs_devices); -mutex_unlock (&uuid_mutex); -if (error) +if (error) { +mutex_unlock (&uuid_mutex); goto error_fs_info; + } fs_info->fs_devices = fs_devices; -mutex_lock (&uuid_mutex); error =btrfs_open_devices (fs_devices, mode, fs_type);mutex_unlock (&uuid_mutex);if (error)
Greg Kroah-Hartman
maintainerPATCH 152/197 drm/i915/lpe: Mark LPE audio runtime pm as "no callbacks"
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chris Wilson <[email protected]>
commit 75eef0f1ed478284911b8723a5bdb659499a7aac upstream.
The LPE audio is a child device of i915, it is powered up and down alongside the igfx and presents no independent runtime interface. This aptly fulfils the description of a "No-Callback" Device, so mark it thus.
Testcase: igt/pm_rpm/basic-pci-d3-state
Testcase: igt/pm_rpm/basic-rte
--- drivers/gpu/drm/i915/intel_lpe_audio.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_lpe_audio.c +1 -3
@@ -127,9 +127,7 @@ lpe_audio_platdev_create(struct drm_i915return platdev; } -pm_runtime_forbid (&platdev->dev); -pm_runtime_set_active (&platdev->dev); -pm_runtime_enable (&platdev->dev); +pm_runtime_no_callbacks (&platdev->dev);return platdev; }
Greg Kroah-Hartman
maintainerPATCH 153/197 drm/i915: Nuke the LVDS lid notifier
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ville Syrjälä <[email protected]>
commit 05c72e77ccda89ff624108b1b59a0fc43843f343 upstream.
We broke the LVDS notifier resume thing in (presumably) commit e2c8b8701e2d ("drm/i915: Use atomic helpers for suspend, v2.") as we no longer duplicate the current state in the LVDS notifier and thus we never resume it properly either.
Instead of trying to fix it again let's just kill off the lid notifier entirely. None of the machines tested thus far have apparently needed it. Originally the lid notifier was added to work around cases where the VBIOS was clobbering some of the hardware state behind the driver's back, mostly on Thinkpads. We now have a few report of Thinkpads working just fine without the notifier. So maybe it was misdiagnosed originally, or something else has changed (ACPI video stuff perhaps?).
If we do end up finding a machine where the VBIOS is still causing problems I would suggest that we first try setting various bits in the VBIOS scratch registers. There are several to choose from that may instruct the VBIOS to steer clear.
With the notifier gone we'll also stop looking at the panel status in ->detect().
v2: Nuke enum modeset_restore (Rodrigo)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105902
References: https://lists.freedesktop.org/archives/intel-gfx/2018-June/169315.html
References: https://bugs.freedesktop.org/show_bug.cgi?id=21230
--- drivers/gpu/drm/i915/i915_drv.c | 10 -- drivers/gpu/drm/i915/i915_drv.h | 8 -- drivers/gpu/drm/i915/intel_lvds.c | 136 -------------------------------------- 3 files changed, 2 insertions(+), 152 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +0 -10
@@ -919,7 +919,6 @@ static int i915_driver_init_early(structspin_lock_init (&dev_priv->uncore.lock);mutex_init (&dev_priv->sb_lock); -mutex_init (&dev_priv->modeset_restore_lock);mutex_init (&dev_priv->av_mutex);mutex_init (&dev_priv->wm.wm_mutex);mutex_init (&dev_priv->pps_mutex);
@@ -1560,11 +1559,6 @@ static int i915_drm_suspend(struct drm_dpci_power_t opregion_target_state;int error; -/* ignore lid events during suspend */ -mutex_lock (&dev_priv->modeset_restore_lock); - dev_priv->modeset_restore =MODESET_SUSPENDED ; -mutex_unlock (&dev_priv->modeset_restore_lock); -disable_rpm_wakeref_asserts (dev_priv);/* We do a lot of poking in a lot of registers, make sure they work
@@ -1764,10 +1758,6 @@ static int i915_drm_resume(struct drm_deintel_fbdev_set_suspend (dev,FBINFO_STATE_RUNNING , false); -mutex_lock (&dev_priv->modeset_restore_lock); - dev_priv->modeset_restore =MODESET_DONE ; -mutex_unlock (&dev_priv->modeset_restore_lock); -intel_opregion_notify_adapter (dev_priv,PCI_D0 );enable_rpm_wakeref_asserts (dev_priv);
drivers/gpu/drm/i915/i915_drv.h +0 -8
@@ -1003,12 +1003,6 @@ struct i915_gem_mm {#define I915_ENGINE_DEAD_TIMEOUT (4 * HZ)/* Seqno, head and subunits dead */ #define I915_SEQNO_DEAD_TIMEOUT (12 * HZ)/* Seqno dead with active head */ -enum modeset_restore { -MODESET_ON_LID_OPEN , -MODESET_DONE , -MODESET_SUSPENDED , -}; -#define DP_AUX_A 0x40#define DP_AUX_B 0x10#define DP_AUX_C 0x20
@@ -1740,8 +1734,6 @@ struct drm_i915_private {unsigned long quirks; -enum modeset_restore modeset_restore; -struct mutex modeset_restore_lock;struct drm_atomic_state *modeset_restore_state;struct drm_modeset_acquire_ctx reset_ctx;
drivers/gpu/drm/i915/intel_lvds.c +2 -134
169 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 154/197 drm/i915: Increase LSPCON timeout
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Fredrik Schön <[email protected]>
commit 299c2a904b1e8d5096d4813df6371357d97a6cd1 upstream.
100 ms is not enough time for the LSPCON adapter on Intel NUC devices to settle. This causes dropped display modes at boot or screen reconfiguration. Empirical testing can reproduce the error up to a timeout of 190 ms. Basic boot and stress testing at 200 ms has not (yet) failed.
Increase timeout to 400 ms to get some margin of error.
Changes from v1:
The initial suggestion of 1000 ms was lowered due to concerns about delaying valid timeout cases.
Update patch metadata.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107503
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1570392
--- drivers/gpu/drm/i915/intel_lspcon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/i915/intel_lspcon.c +1 -1
@@ -74,7 +74,7 @@ static enum drm_lspcon_mode lspcon_wait_DRM_DEBUG_KMS ("Waiting for LSPCON mode %s to settle\n" ,lspcon_mode_name (mode)); -wait_for ((current_mode =lspcon_get_current_mode (lspcon)) == mode,100 ); +wait_for ((current_mode =lspcon_get_current_mode (lspcon)) == mode,400 );if (current_mode != mode)DRM_ERROR ("LSPCON mode hasn't settled\n" );
Greg Kroah-Hartman
maintainerPATCH 155/197 drm/i915: Free write_buf that we allocated with kzalloc.
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rodrigo Vivi <[email protected]>
commit 1b1b1162745e5f9e5c6c095afc8081df3edabc50 upstream.
We use kzalloc to allocate the write_buf that we use for i2c transfer on hdcp write. But it seems that we are forgetting to free the memory that is not needed after i2c transfer is completed.
--- drivers/gpu/drm/i915/intel_hdmi.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/intel_hdmi.c +6 -2
@@ -933,8 +933,12 @@ static int intel_hdmi_hdcp_write(struct ret =i2c_transfer (adapter, &msg,1 );if (ret ==1 ) -return 0 ; -return ret >=0 ? -EIO : ret; + ret =0 ; +else if (ret >=0 ) + ret = -EIO ; + +kfree (write_buf); +return ret; } static
Greg Kroah-Hartman
maintainerPATCH 156/197 drm/amdgpu: update uvd_v6_0_ring_vm_funcs to use new nop packet
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit afb1436c7b44ab928e6369a4d48e3abb8215241e upstream.
Was missed when updating the uvd 6 module.
--- drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +1 -2
@@ -1569,7 +1569,6 @@ static const struct amdgpu_ring_funcs uvstatic const struct amdgpu_ring_funcs uvd_v6_0_ring_vm_funcs = { .type =AMDGPU_RING_TYPE_UVD , .align_mask =0xf , - .nop =PACKET0 (mmUVD_NO_OP,0 ), .support_64bit_ptrs = false, .get_rptr = uvd_v6_0_ring_get_rptr, .get_wptr = uvd_v6_0_ring_get_wptr,
@@ -1587,7 +1586,7 @@ static const struct amdgpu_ring_funcs uv .emit_hdp_flush = uvd_v6_0_ring_emit_hdp_flush, .test_ring = uvd_v6_0_ring_test_ring, .test_ib = amdgpu_uvd_ring_test_ib, - .insert_nop = amdgpu_ring_insert_nop, + .insert_nop = uvd_v6_0_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .begin_use = amdgpu_uvd_ring_begin_use, .end_use = amdgpu_uvd_ring_end_use,
Greg Kroah-Hartman
maintainerPATCH 157/197 drm/amdgpu: fix a reversed condition
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rex Zhu <[email protected]>
commit ccf9ef0b0d10434dec5046bcfc4e834a7b1830fd upstream.
This test was reversed so it would end up leading to vddnb value can't be read via hwmon on APU.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +1 -1
@@ -1157,7 +1157,7 @@ static ssize_t amdgpu_hwmon_show_vddnb(sint r, size =sizeof (vddnb);/* only APUs have vddnb */ -if (adev->flags &AMD_IS_APU ) +if (!(adev->flags &AMD_IS_APU ))return -EINVAL ;/* Can't get voltage when the card is off */
Greg Kroah-Hartman
maintainerPATCH 158/197 drm/amdgpu: Fix RLC safe mode test in gfx_v9_0_enter_rlc_safe_mode
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michel Dänzer <[email protected]>
commit 226127a67e31a9518d9516d3e4890759b379d874 upstream.
We were testing the register offset, instead of the value stored in the register, therefore always timing out the loop.
This reduces suspend time of the system in the bug report below by ~600 ms.
Cc: [email protected]
Bugzilla: https://bugs.freedesktop.org/107277
--- drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +1 -1
@@ -3433,7 +3433,7 @@ static void gfx_v9_0_enter_rlc_safe_mode/* wait for RLC_SAFE_MODE */ for (i =0 ; i < adev->usec_timeout; i++) { -if (!REG_GET_FIELD (SOC15_REG_OFFSET (GC ,0 , mmRLC_SAFE_MODE),RLC_SAFE_MODE ,CMD )) +if (!REG_GET_FIELD (RREG32_SOC15 (GC ,0 , mmRLC_SAFE_MODE),RLC_SAFE_MODE ,CMD ))break ;udelay (1 ); }
Greg Kroah-Hartman
maintainerPATCH 159/197 drm/amd/pp: Convert voltage unit in mV*4 to mV on CZ/ST
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rex Zhu <[email protected]>
commit 8a50bb47a863c3cb8950a2e810448c9a82a9d446 upstream.
the voltage showed in debugfs and hwmon should be in mV
--- drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c +3 -2
@@ -244,6 +244,7 @@ static int smu8_initialize_dpm_defaults(return 0 ; } +/* convert form 8bit vid to real voltage in mV*4 */ static uint32_t smu8_convert_8Bit_index_to_voltage (struct pp_hwmgr *hwmgr,uint16_t voltage) {
@@ -1702,13 +1703,13 @@ static int smu8_read_sensor(struct pp_hwcase AMDGPU_PP_SENSOR_VDDNB : tmp = (cgs_read_ind_register (hwmgr->device,CGS_IND_REG__SMC , ixSMUSVI_NB_CURRENTVID) &CURRENT_NB_VID_MASK ) >>CURRENT_NB_VID__SHIFT ; - vddnb =smu8_convert_8Bit_index_to_voltage (hwmgr, tmp); + vddnb =smu8_convert_8Bit_index_to_voltage (hwmgr, tmp) /4 ; *((uint32_t *)value) = vddnb;return 0 ;case AMDGPU_PP_SENSOR_VDDGFX : tmp = (cgs_read_ind_register (hwmgr->device,CGS_IND_REG__SMC , ixSMUSVI_GFX_CURRENTVID) &CURRENT_GFX_VID_MASK ) >>CURRENT_GFX_VID__SHIFT ; - vddgfx =smu8_convert_8Bit_index_to_voltage (hwmgr, (u16 )tmp); + vddgfx =smu8_convert_8Bit_index_to_voltage (hwmgr, (u16 )tmp) /4 ; *((uint32_t *)value) = vddgfx;return 0 ;case AMDGPU_PP_SENSOR_UVD_VCLK :
Greg Kroah-Hartman
maintainerPATCH 160/197 drm/amd/powerplay: fixed uninitialized value
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Evan Quan <[email protected]>
commit 1ce0688f3f6a9e9d34ae66bf779d54855def7bec upstream.
The 'result' is not initialized correctly. It causes the API return an error code even on success.
--- drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c +1 -1
@@ -490,7 +490,7 @@ static int vega12_get_number_dpm_level(sstatic int vega12_get_dpm_frequency_by_index (struct pp_hwmgr *hwmgr,PPCLK_e clkID,uint32_t index,uint32_t *clock) { -int result; +int result =0 ;/* *SMU expects the Clock ID to be in the top 16 bits.
Greg Kroah-Hartman
maintainerPATCH 161/197 drm/amd/pp/Polaris12: Fix a chunk of registers missed to program
In reply to Greg Kroah-Hartman4.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rex Zhu <[email protected]>
commit 2d227ec2c11c568910299e8f913bac2dda47397c upstream.
DIDTConfig_Polaris12[] table missed a big chunk of data.
Pointed by aidan.fabius <[email protected]>
--- drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c | 43 +++++++++++++++++++ 1 file changed, 43 insertions(+)
drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c +43 -0
@@ -403,6 +403,49 @@ static const struct gpu_pt_config_reg DI { ixDIDT_SQ_CTRL1, DIDT_SQ_CTRL1__MAX_POWER_MASK, DIDT_SQ_CTRL1__MAX_POWER__SHIFT, 0xffff, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_SQ_CTRL_OCP, DIDT_SQ_CTRL_OCP__UNUSED_0_MASK, DIDT_SQ_CTRL_OCP__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL_OCP, DIDT_SQ_CTRL_OCP__OCP_MAX_POWER_MASK, DIDT_SQ_CTRL_OCP__OCP_MAX_POWER__SHIFT, 0xffff, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_SQ_CTRL2, DIDT_SQ_CTRL2__MAX_POWER_DELTA_MASK, DIDT_SQ_CTRL2__MAX_POWER_DELTA__SHIFT, 0x3853, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL2, DIDT_SQ_CTRL2__UNUSED_0_MASK, DIDT_SQ_CTRL2__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL2, DIDT_SQ_CTRL2__SHORT_TERM_INTERVAL_SIZE_MASK, DIDT_SQ_CTRL2__SHORT_TERM_INTERVAL_SIZE__SHIFT, 0x005a, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL2, DIDT_SQ_CTRL2__UNUSED_1_MASK, DIDT_SQ_CTRL2__UNUSED_1__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL2, DIDT_SQ_CTRL2__LONG_TERM_INTERVAL_RATIO_MASK, DIDT_SQ_CTRL2__LONG_TERM_INTERVAL_RATIO__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL2, DIDT_SQ_CTRL2__UNUSED_2_MASK, DIDT_SQ_CTRL2__UNUSED_2__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_SQ_STALL_CTRL, DIDT_SQ_STALL_CTRL__DIDT_STALL_CTRL_ENABLE_MASK, DIDT_SQ_STALL_CTRL__DIDT_STALL_CTRL_ENABLE__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_STALL_CTRL, DIDT_SQ_STALL_CTRL__DIDT_STALL_DELAY_HI_MASK, DIDT_SQ_STALL_CTRL__DIDT_STALL_DELAY_HI__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_STALL_CTRL, DIDT_SQ_STALL_CTRL__DIDT_STALL_DELAY_LO_MASK, DIDT_SQ_STALL_CTRL__DIDT_STALL_DELAY_LO__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_STALL_CTRL, DIDT_SQ_STALL_CTRL__DIDT_HI_POWER_THRESHOLD_MASK, DIDT_SQ_STALL_CTRL__DIDT_HI_POWER_THRESHOLD__SHIFT, 0x0ebb, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_STALL_CTRL, DIDT_SQ_STALL_CTRL__UNUSED_0_MASK, DIDT_SQ_STALL_CTRL__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_SQ_TUNING_CTRL, DIDT_SQ_TUNING_CTRL__DIDT_TUNING_ENABLE_MASK, DIDT_SQ_TUNING_CTRL__DIDT_TUNING_ENABLE__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_TUNING_CTRL, DIDT_SQ_TUNING_CTRL__MAX_POWER_DELTA_HI_MASK, DIDT_SQ_TUNING_CTRL__MAX_POWER_DELTA_HI__SHIFT, 0x3853, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_TUNING_CTRL, DIDT_SQ_TUNING_CTRL__MAX_POWER_DELTA_LO_MASK, DIDT_SQ_TUNING_CTRL__MAX_POWER_DELTA_LO__SHIFT, 0x3153, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_TUNING_CTRL, DIDT_SQ_TUNING_CTRL__UNUSED_0_MASK, DIDT_SQ_TUNING_CTRL__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__DIDT_CTRL_EN_MASK, DIDT_SQ_CTRL0__DIDT_CTRL_EN__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__USE_REF_CLOCK_MASK, DIDT_SQ_CTRL0__USE_REF_CLOCK__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__PHASE_OFFSET_MASK, DIDT_SQ_CTRL0__PHASE_OFFSET__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__DIDT_CTRL_RST_MASK, DIDT_SQ_CTRL0__DIDT_CTRL_RST__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__DIDT_CLK_EN_OVERRIDE_MASK, DIDT_SQ_CTRL0__DIDT_CLK_EN_OVERRIDE__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__DIDT_MAX_STALLS_ALLOWED_HI_MASK, DIDT_SQ_CTRL0__DIDT_MAX_STALLS_ALLOWED_HI__SHIFT, 0x0010, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__DIDT_MAX_STALLS_ALLOWED_LO_MASK, DIDT_SQ_CTRL0__DIDT_MAX_STALLS_ALLOWED_LO__SHIFT, 0x0010, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_SQ_CTRL0, DIDT_SQ_CTRL0__UNUSED_0_MASK, DIDT_SQ_CTRL0__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_TD_WEIGHT0_3, DIDT_TD_WEIGHT0_3__WEIGHT0_MASK, DIDT_TD_WEIGHT0_3__WEIGHT0__SHIFT, 0x000a, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TD_WEIGHT0_3, DIDT_TD_WEIGHT0_3__WEIGHT1_MASK, DIDT_TD_WEIGHT0_3__WEIGHT1__SHIFT, 0x0010, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TD_WEIGHT0_3, DIDT_TD_WEIGHT0_3__WEIGHT2_MASK, DIDT_TD_WEIGHT0_3__WEIGHT2__SHIFT, 0x0017, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TD_WEIGHT0_3, DIDT_TD_WEIGHT0_3__WEIGHT3_MASK, DIDT_TD_WEIGHT0_3__WEIGHT3__SHIFT, 0x002f, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_TD_WEIGHT4_7, DIDT_TD_WEIGHT4_7__WEIGHT4_MASK, DIDT_TD_WEIGHT4_7__WEIGHT4__SHIFT, 0x0046, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TD_WEIGHT4_7, DIDT_TD_WEIGHT4_7__WEIGHT5_MASK, DIDT_TD_WEIGHT4_7__WEIGHT5__SHIFT, 0x005d, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TD_WEIGHT4_7, DIDT_TD_WEIGHT4_7__WEIGHT6_MASK, DIDT_TD_WEIGHT4_7__WEIGHT6__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TD_WEIGHT4_7, DIDT_TD_WEIGHT4_7__WEIGHT7_MASK, DIDT_TD_WEIGHT4_7__WEIGHT7__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_TD_CTRL1, DIDT_TD_CTRL1__MIN_POWER_MASK, DIDT_TD_CTRL1__MIN_POWER__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TD_CTRL1, DIDT_TD_CTRL1__MAX_POWER_MASK, DIDT_TD_CTRL1__MAX_POWER__SHIFT, 0xffff, GPU_CONFIGREG_DIDT_IND }, + + { ixDIDT_TD_CTRL_OCP, DIDT_TD_CTRL_OCP__UNUSED_0_MASK, DIDT_TD_CTRL_OCP__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TD_CTRL_OCP, DIDT_TD_CTRL_OCP__OCP_MAX_POWER_MASK, DIDT_TD_CTRL_OCP__OCP_MAX_POWER__SHIFT, 0x00ff, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TD_CTRL2, DIDT_TD_CTRL2__MAX_POWER_DELTA_MASK, DIDT_TD_CTRL2__MAX_POWER_DELTA__SHIFT, 0x3fff, GPU_CONFIGREG_DIDT_IND },