5.18.13-rc1 review
Subsystems (3)
Greg Kroah-Hartman
maintainercover 5.18.13-rc1 review
This is the start of the stable review cycle for the 5.18.13 release. There are 231 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Thu, 21 Jul 2022 11:43:40 +0000. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.18.13-rc1.gz
or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.18.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 5.18.13-rc1
Juergen Gross <[email protected]>
x86/pat: Fix x86_has_pat_wp()
Pali Rohár <[email protected]>
serial: mvebu-uart: correctly report configured baudrate value
Ilpo Järvinen <[email protected]>
serial: 8250: Fix PM usage_count for console handover
Ilpo Järvinen <[email protected]>
serial: pl011: UPSTAT_AUTORTS requires .throttle/unthrottle
Ilpo Järvinen <[email protected]>
serial: stm32: Clear prev values before setting RTS delays
Dorian Rudolph <[email protected]>
power: supply: core: Fix boundary conditions in interpolation
Yi Yang <[email protected]>
serial: 8250: fix return error code in serial8250_request_std_resource()
Mario Limonciello <[email protected]>
ACPI: CPPC: Fix enabling CPPC on AMD systems with shared memory
Tony Krowiak <[email protected]>
s390/ap: fix error handling in __verify_queue_reservations()
Yangxi Xiang <[email protected]>
vt: fix memory overlapping when deleting chars in the buffer
Hans de Goede <[email protected]>
ACPI: video: Fix acpi_video_handles_brightness_key_presses()
Linus Torvalds <[email protected]>
signal handling: don't use BUG_ON() for debugging
Jason A. Donenfeld <[email protected]>
wireguard: selftests: always call kernel makefile
Jason A. Donenfeld <[email protected]>
wireguard: selftests: set fake real time in init
Keith Busch <[email protected]>
nvme: use struct group for generic command dwords
Keith Busch <[email protected]>
nvme-pci: phison e16 has bogus namespace ids
Egor Vorontsov <[email protected]>
ALSA: usb-audio: Add quirk for Fiero SC-01 (fw v1.0.0)
Egor Vorontsov <[email protected]>
ALSA: usb-audio: Add quirk for Fiero SC-01
John Veness <[email protected]>
ALSA: usb-audio: Add quirks for MacroSilicon MS2100/MS2106 devices
Srinivas Neeli <[email protected]>
Revert "can: xilinx_can: Limit CANFD brp to 2"
Gabriel Fernandez <[email protected]>
ARM: dts: stm32: use the correct clock source for CEC on stm32mp151
Linus Walleij <[email protected]>
soc: ixp4xx/npe: Fix unused match warning
Juergen Gross <[email protected]>
x86: Clear .brk area at early boot
Jacky Bai <[email protected]>
pinctrl: imx: Add the zero base flag for imx93
Stafford Horne <[email protected]>
irqchip: or1k-pic: Undefine mask_ack for level triggered hardware
Charles Keepax <[email protected]>
ASoC: madera: Fix event generation for rate controls
Charles Keepax <[email protected]>
ASoC: madera: Fix event generation for OUT1 demux
Charles Keepax <[email protected]>
ASoC: cs47l15: Fix event generation for low power mux control
Charles Keepax <[email protected]>
ASoC: cs35l41: Add ASP TX3/4 source to register patch
Charles Keepax <[email protected]>
ASoC: dapm: Initialise kcontrol data for mux/demux controls
Shuming Fan <[email protected]>
ASoC: rt711-sdca: fix kernel NULL pointer dereference when IO error
Charles Keepax <[email protected]>
ASoC: cs35l41: Correct some control names
Charles Keepax <[email protected]>
ASoC: wm5110: Fix DRE control
Charles Keepax <[email protected]>
ASoC: wm_adsp: Fix event for preloader
Hans de Goede <[email protected]>
ASoC: Intel: bytcr_wm5102: Fix GPIO related probe-ordering problem
Mark Brown <[email protected]>
ASoC: wcd938x: Fix event generation for some controls
Mark Brown <[email protected]>
ASoC: wcd9335: Fix spurious event generation
Yassine Oudjana <[email protected]>
ASoC: wcd9335: Remove RX channel from old list before adding it to a new one
Peter Ujfalusi <[email protected]>
ASoC: SOF: Intel: hda-loader: Clarify the cl_dsp_init() flow
Peter Ujfalusi <[email protected]>
ASoC: SOF: Intel: hda-loader: Make sure that the fw load sequence is followed
Peter Ujfalusi <[email protected]>
ASoC: SOF: Intel: hda-dsp: Expose hda_dsp_core_power_up()
Pierre-Louis Bossart <[email protected]>
ASoC: codecs: rt700/rt711/rt711-sdca: initialize workqueues in probe
Pierre-Louis Bossart <[email protected]>
ASoC: rt7*-sdw: harden jack_detect_handler
Pierre-Louis Bossart <[email protected]>
ASoC: rt711: fix calibrate mutex initialization
Pierre-Louis Bossart <[email protected]>
ASoC: Intel: sof_sdw: handle errors on card registration
Pierre-Louis Bossart <[email protected]>
ASoC: rt711-sdca-sdw: fix calibrate mutex initialization
Pierre-Louis Bossart <[email protected]>
ASoC: Realtek/Maxim SoundWire codecs: disable pm_runtime on remove
Haowen Bai <[email protected]>
pinctrl: aspeed: Fix potential NULL dereference in aspeed_pinmux_set_mux()
Mark Brown <[email protected]>
ASoC: ops: Fix off by one in range control validation
Jianglei Nie <[email protected]>
net: sfp: fix memory leak in sfp_probe()
Ruozhu Li <[email protected]>
nvme: fix regression when disconnect a recovering ctrl
Sagi Grimberg <[email protected]>
nvme-tcp: always fail a request when sending it failed
Michael Walle <[email protected]>
NFC: nxp-nci: don't print header length mismatch on i2c error
Hangyu Hua <[email protected]>
net: tipc: fix possible refcount leak in tipc_sk_create()
Javier Martinez Canillas <[email protected]>
fbdev: Disable sysfb device registration when removing conflicting FBs
Javier Martinez Canillas <[email protected]>
firmware: sysfb: Add sysfb_disable() helper function
Javier Martinez Canillas <[email protected]>
firmware: sysfb: Make sysfb_create_simplefb() return a pdev pointer
Kai-Heng Feng <[email protected]>
platform/x86: hp-wmi: Ignore Sanitization Mode event
Mark Pearson <[email protected]>
platform/x86: thinkpad_acpi: do not use PSC mode on Intel platforms
Mark Pearson <[email protected]>
platform/x86: thinkpad-acpi: profile capabilities as integer
Liang He <[email protected]>
cpufreq: pmac32-cpufreq: Fix refcount leak bug
John Garry <[email protected]>
scsi: hisi_sas: Limit max hw sectors for v3 HW
Florian Westphal <[email protected]>
netfilter: br_netfilter: do not skip all hooks with 0 priority
Florian Westphal <[email protected]>
netfilter: nf_tables: avoid skb access on nf_stolen
Stephan Gerhold <[email protected]>
virtio_mmio: Restore guest page size on resume
Stephan Gerhold <[email protected]>
virtio_mmio: Add missing PM calls to freeze/restore
Gayatri Kammela <[email protected]>
platform/x86: intel/pmc: Add Alder Lake N support to PMC core driver
Parav Pandit <[email protected]>
vduse: Tie vduse mgmtdev and its device
Eli Cohen <[email protected]>
vdpa/mlx5: Initialize CVQ vringh only once
Nathan Lynch <[email protected]>
powerpc/xive/spapr: correct bitmap allocation size
Namjae Jeon <[email protected]>
ksmbd: use SOCK_NONBLOCK type for kernel_accept()
Israel Rukshin <[email protected]>
nvme: fix block device naming collision
Bjorn Andersson <[email protected]>
scsi: ufs: core: Drop loglevel of WriteBoost message
Ming Lei <[email protected]>
scsi: megaraid: Clear READ queue map's nr_queues
Vasily Gorbik <[email protected]>
s390/nospec: build expoline.o for modules_prepare target
Jiri Slaby <[email protected]>
tty: use new tty_insert_flip_string_and_push_buffer() in pty_write()
Jiri Slaby <[email protected]>
tty: extract tty_flip_buffer_commit() from tty_flip_buffer_push()
Marc Kleine-Budde <[email protected]>
tee: tee_get_drvdata(): fix description of return value
Bryan O'Donoghue <[email protected]>
ASoC: dt-bindings: Fix description for msm8916
Linus Walleij <[email protected]>
ARM: 9212/1: domain: Modify Kconfig help text
Linus Walleij <[email protected]>
ARM: 9211/1: domain: drop modify_domain()
Peter Zijlstra <[email protected]>
x86/cpu/amd: Add Spectral Chicken
Peter Zijlstra <[email protected]>
x86/xen: Rename SYS* entry points
Peter Zijlstra <[email protected]>
objtool: Update Retpoline validation
Alexandre Chartre <[email protected]>
x86/bugs: Report AMD retbleed vulnerability
Kim Phillips <[email protected]>
x86/sev: Avoid using __x86_return_thunk
Peter Zijlstra <[email protected]>
x86/kvm: Fix SETcc emulation for return thunks
Muchun Song <[email protected]>
mm: sysctl: fix missing numa_stat when !CONFIG_HUGETLB_PAGE
Tariq Toukan <[email protected]>
net/tls: Check for errors in tls_device_init
Vitaly Kuznetsov <[email protected]>
KVM: x86: Fully initialize 'struct kvm_lapic_irq' in kvm_pv_kick_cpu_op()
Chia-Lin Kao (AceLan) <[email protected]>
net: atlantic: remove aq_nic_deinit() when resume
Chia-Lin Kao (AceLan) <[email protected]>
net: atlantic: remove deep parameter on suspend/resume functions
Íñigo Huguet <[email protected]>
sfc: fix kernel panic when creating VF
Andrea Mayer <[email protected]>
seg6: bpf: fix skb checksum in bpf_push_seg6_encap()
Andrea Mayer <[email protected]>
seg6: fix skb checksum in SRv6 End.B6 and End.B6.Encaps behaviors
Andrea Mayer <[email protected]>
seg6: fix skb checksum evaluation in SRH encapsulation/insertion
Xiubo Li <[email protected]>
netfs: do not unlock and put the folio twice
Íñigo Huguet <[email protected]>
sfc: fix use after free when disabling sriov
Steve French <[email protected]>
smb3: workaround negprot bug in some Samba servers
Michel Dänzer <[email protected]>
drm/amd/display: Ensure valid event timestamp for cursor-only commits
Yefim Barashkin <[email protected]>
drm/amd/pm: Prevent divide by zero
Mario Kleiner <[email protected]>
drm/amd/display: Only use depth 36 bpp linebuffers on DCN display engines.
Prike Liang <[email protected]>
drm/amdkfd: correct the MEC atomic support firmware checking for GC 10.3.7
Fangzhi Zuo <[email protected]>
drm/amd/display: Ignore First MST Sideband Message Return Error
Jianglei Nie <[email protected]>
ima: Fix potential memory leak in ima_init_crypto()
Coiby Xu <[email protected]>
ima: force signature verification when CONFIG_KEXEC_SIG is configured
Dan Carpenter <[email protected]>
net: stmmac: fix leaks in probe
Liang He <[email protected]>
net: ftgmac100: Hold reference returned by of_get_child_by_name()
Kuniyuki Iwashima <[email protected]>
nexthop: Fix data-races around nexthop_compat_mode.
Kuniyuki Iwashima <[email protected]>
ipv4: Fix data-races around sysctl_ip_dynaddr.
Kuniyuki Iwashima <[email protected]>
tcp: Fix a data-race around sysctl_tcp_ecn_fallback.
Kuniyuki Iwashima <[email protected]>
tcp: Fix data-races around sysctl_tcp_ecn.
Kuniyuki Iwashima <[email protected]>
raw: Fix a data-race around sysctl_raw_l3mdev_accept.
Kuniyuki Iwashima <[email protected]>
icmp: Fix a data-race around sysctl_icmp_ratemask.
Kuniyuki Iwashima <[email protected]>
icmp: Fix a data-race around sysctl_icmp_ratelimit.
Kuniyuki Iwashima <[email protected]>
icmp: Fix a data-race around sysctl_icmp_errors_use_inbound_ifaddr.
Kuniyuki Iwashima <[email protected]>
icmp: Fix a data-race around sysctl_icmp_ignore_bogus_error_responses.
Kuniyuki Iwashima <[email protected]>
icmp: Fix a data-race around sysctl_icmp_echo_ignore_broadcasts.
Kuniyuki Iwashima <[email protected]>
icmp: Fix data-races around sysctl_icmp_echo_enable_probe.
Kuniyuki Iwashima <[email protected]>
icmp: Fix a data-race around sysctl_icmp_echo_ignore_all.
Kuniyuki Iwashima <[email protected]>
tcp: Fix a data-race around sysctl_max_tw_buckets.
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data-races in proc_dointvec_ms_jiffies().
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data-races in proc_dou8vec_minmax().
Pavan Chebbi <[email protected]>
bnxt_en: Fix bnxt_refclk_read()
Vikas Gupta <[email protected]>
bnxt_en: fix livepatch query
Michael Chan <[email protected]>
bnxt_en: Fix bnxt_reinit_after_abort() code path
Kashyap Desai <[email protected]>
bnxt_en: reclaim max resources if sriov enable fails
Andrzej Hajda <[email protected]>
drm/i915/selftests: fix subtraction overflow bug
Chris Wilson <[email protected]>
drm/i915/gt: Serialize TLB invalidates with GT resets
Chris Wilson <[email protected]>
drm/i915/gt: Serialize GRDOM access between multiple engine resets
Matthew Auld <[email protected]>
drm/i915/ttm: fix sg_table construction
Dan Carpenter <[email protected]>
drm/i915/selftests: fix a couple IS_ERR() vs NULL tests
Douglas Anderson <[email protected]>
tracing: Fix sleeping while atomic in kdb ftdump
Paul M Stillwell Jr <[email protected]>
ice: change devlink code to read NVM in blocks
Paul M Stillwell Jr <[email protected]>
ice: handle E822 generic device ID in PLDM header
Yevhen Orlov <[email protected]>
net: marvell: prestera: fix missed deinit sequence
Jeff Layton <[email protected]>
lockd: fix nlm_close_files
Jeff Layton <[email protected]>
lockd: set fl_owner when unlocking files
Chuck Lever <[email protected]>
NFSD: Decode NFSv4 birth time attribute
Paolo Abeni <[email protected]>
mptcp: fix subflow traversal at disconnect time
Demi Marie Obenour <[email protected]>
xen/gntdev: Ignore failure to unmap INVALID_GRANT_HANDLE
Dan Carpenter <[email protected]>
drm/i915/gvt: IS_ERR() vs NULL bug in intel_gvt_update_reg_whitelist()
Anup Patel <[email protected]>
RISC-V: KVM: Fix SRCU deadlock caused by kvm_riscv_check_vcpu_requests()
Pablo Neira Ayuso <[email protected]>
netfilter: nf_tables: replace BUG_ON by element length check
Eric Dumazet <[email protected]>
vlan: fix memory leak in vlan_newlink()
Baowen Zheng <[email protected]>
nfp: fix issue of skb segments exceeds descriptor limitation
Pablo Neira Ayuso <[email protected]>
netfilter: nf_log: incorrect offset to network header
William Zhang <[email protected]>
arm64: dts: broadcom: bcm4908: Fix cpu node for smp boot
William Zhang <[email protected]>
arm64: dts: broadcom: bcm4908: Fix timer node for BCM4906 SoC
Michal Suchanek <[email protected]>
ARM: dts: sunxi: Fix SPI NOR campatible on Orange Pi Zero
Ryan Wanner <[email protected]>
ARM: dts: at91: sama5d2: Fix typo in i2s1 node
Kuniyuki Iwashima <[email protected]>
ipv4: Fix a data-race around sysctl_fib_sync_mem.
Kuniyuki Iwashima <[email protected]>
icmp: Fix data-races around sysctl.
Kuniyuki Iwashima <[email protected]>
cipso: Fix data-races around sysctl.
Kuniyuki Iwashima <[email protected]>
net: Fix data-races around sysctl_mem.
Kuniyuki Iwashima <[email protected]>
inetpeer: Fix data-races around sysctl.
Kuniyuki Iwashima <[email protected]>
tcp: Fix a data-race around sysctl_tcp_max_orphans.
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data races in proc_dointvec_jiffies().
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data races in proc_doulongvec_minmax().
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data races in proc_douintvec_minmax().
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data races in proc_dointvec_minmax().
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data races in proc_douintvec().
Kuniyuki Iwashima <[email protected]>
sysctl: Fix data races in proc_dointvec().
Pavel Skripkin <[email protected]>
net: ocelot: fix wrong time_after usage
Siddharth Vadapalli <[email protected]>
net: ethernet: ti: am65-cpsw: Fix devlink port register sequence
Jon Hunter <[email protected]>
net: stmmac: dwc-qos: Disable split header for Tegra194
Florian Westphal <[email protected]>
netfilter: conntrack: fix crash due to confirmed bit load reordering
Florian Westphal <[email protected]>
netfilter: conntrack: remove the percpu dying list
Florian Westphal <[email protected]>
netfilter: conntrack: include ecache dying list in dumps
Florian Westphal <[email protected]>
netfilter: ecache: use dedicated list for event redelivery
Florian Westphal <[email protected]>
netfilter: conntrack: split inner loop of list dumping to own function
Florian Westphal <[email protected]>
netfilter: ecache: move to separate structure
Peter Ujfalusi <[email protected]>
ASoC: Intel: Skylake: Correct the handling of fmt_config flexible array
Peter Ujfalusi <[email protected]>
ASoC: Intel: Skylake: Correct the ssp rate discovery in skl_get_ssp_clks()
Hector Martin <[email protected]>
ASoC: tas2764: Fix amp gain register offset & default
Hector Martin <[email protected]>
ASoC: tas2764: Correct playback volume range
Martin Povišer <[email protected]>
ASoC: tas2764: Fix and extend FSYNC polarity handling
Martin Povišer <[email protected]>
ASoC: tas2764: Add post reset delays
Francesco Dolcini <[email protected]>
ASoC: sgtl5000: Fix noise on shutdown/remove
Huaxin Lu <[email protected]>
ima: Fix a potential integer overflow in ima_appraise_measurement
Daniele Ceraolo Spurio <[email protected]>
drm/i915/guc: ADL-N should use the same GuC FW as ADL-S
Hangyu Hua <[email protected]>
drm/i915: fix a possible refcount leak in intel_dp_add_mst_connector()
Alex Deucher <[email protected]>
drm/amdgpu/display: disable prefer_shadow for generic fb helpers
Alex Deucher <[email protected]>
drm/amdgpu: keep fbdev buffers pinned during suspend
Maxim Mikityanskiy <[email protected]>
net/mlx5e: Ring the TX doorbell on DMA errors
Gal Pressman <[email protected]>
net/mlx5e: Fix capability check for updating vnic env counters
Roi Dayan <[email protected]>
net/mlx5e: CT: Use own workqueue instead of mlx5e priv
Paul Blakey <[email protected]>
net/mlx5e: Fix enabling sriov while tc nic rules are offloaded
Tariq Toukan <[email protected]>
net/mlx5e: kTLS, Fix build time constant test in RX
Tariq Toukan <[email protected]>
net/mlx5e: kTLS, Fix build time constant test in TX
Zhen Lei <[email protected]>
ARM: 9210/1: Mark the FDT_FIXED sections as shareable
Ard Biesheuvel <[email protected]>
ARM: 9209/1: Spectre-BHB: avoid pr_info() every time a CPU comes out of idle
Cristian Ciocaltea <[email protected]>
spi: amd: Limit max transfer and message size
Conor Dooley <[email protected]>
riscv: dts: microchip: hook up the mpfs' l2cache
Kris Bahnsen <[email protected]>
ARM: dts: imx6qdl-ts7970: Fix ngpio typo and count
Sean Anderson <[email protected]>
arm64: dts: ls1028a: Update SFP node to include clock
Serge Semin <[email protected]>
reset: Fix devm bulk optional exclusive control getter
Xiu Jianfeng <[email protected]>
Revert "evm: Fix memleak in init_desc"
Geert Uytterhoeven <[email protected]>
sh: convert nommu io{re,un}map() to static inline functions
Ryusuke Konishi <[email protected]>
nilfs2: fix incorrect masking of permission flags for symlinks
Dave Chinner <[email protected]>
fs/remap: constrain dedupe of EOF blocks
Dmitry Osipenko <[email protected]>
drm/panfrost: Fix shrinker list corruption by madvise IOCTL
Dmitry Osipenko <[email protected]>
drm/panfrost: Put mapping instead of shmem obj on panfrost_mmu_map_fault_addr() error
Christoph Hellwig <[email protected]>
btrfs: zoned: fix a leaked bioc in read_zone_info
Filipe Manana <[email protected]>
btrfs: return -EAGAIN for NOWAIT dio reads/writes on compressed and inline extents
Tejun Heo <[email protected]>
cgroup: Use separate src/dst nodes when preloading css_sets for migration
Felix Fietkau <[email protected]>
wifi: mac80211: fix queue selection for mesh/OCB interfaces
Ard Biesheuvel <[email protected]>
ARM: 9214/1: alignment: advance IT state after emulating Thumb instruction
Dmitry Osipenko <[email protected]>
ARM: 9213/1: Print message about disabled Spectre workarounds only once
Nicolas Dichtel <[email protected]>
ip: fix dflt addr selection for connected nexthop
Steven Rostedt (Google) <[email protected]>
net: sock: tracing: Fix sock_exceed_buf_limit not to dereference stale pointer
Zheng Yejian <[email protected]>
tracing/histograms: Fix memory leak problem
Baolin Wang <[email protected]>
mm/damon: use set_huge_pte_at() to make huge pte old
Gowans, James <[email protected]>
mm: split huge PUD on wp_huge_pud fallback
Muchun Song <[email protected]>
mm: sparsemem: fix missing higher order allocation splitting
Axel Rasmussen <[email protected]>
mm: userfaultfd: fix UFFDIO_CONTINUE on fallocated shmem pages
Oleg Nesterov <[email protected]>
fix race between exit_itimers() and /proc/pid/timers
Juergen Gross <[email protected]>
xen/netback: avoid entering xenvif_rx_next_skb() with an empty rx queue
Meng Tang <[email protected]>
ALSA: hda/realtek - Enable the headset-mic on a Xiaomi's laptop
Meng Tang <[email protected]>
ALSA: hda/realtek - Fix headset mic problem for a HP machine with alc221
Jeremy Szu <[email protected]>
ALSA: hda/realtek: fix mute/micmute LEDs for HP machines
Meng Tang <[email protected]>
ALSA: hda/realtek - Fix headset mic problem for a HP machine with alc671
Meng Tang <[email protected]>
ALSA: hda/realtek: Fix headset mic for Acer SF313-51
Meng Tang <[email protected]>
ALSA: hda/conexant: Apply quirk for another HP ProDesk 600 G3 model
Meng Tang <[email protected]>
ALSA: hda - Add fixup for Dell Latitidue E5430
Juergen Gross <[email protected]>
x86/xen: Use clear_bss() for Xen PV guests
Chanho Park <[email protected]>
tty: serial: samsung_tty: set dma burst_size to 1
Bartosz Golaszewski <[email protected]>
gpio: sim: fix the chip_name configfs item
Thinh Nguyen <[email protected]>
usb: dwc3: gadget: Fix event pending check
Linyu Yuan <[email protected]>
usb: typec: add missing uevent when partner support PD
Lucien Buchmann <[email protected]>
USB: serial: ftdi_sio: add Belimo device ids
-------------
Diffstat:
.../devicetree/bindings/sound/qcom,lpass-cpu.yaml | 8 +-
.../driver-api/firmware/other_interfaces.rst | 6 +
Documentation/filesystems/netfs_library.rst | 8 +-
Documentation/networking/ip-sysctl.rst | 4 +-
Makefile | 4 +-
arch/arm/boot/dts/imx6qdl-ts7970.dtsi | 2 +-
arch/arm/boot/dts/sama5d2.dtsi | 2 +-
arch/arm/boot/dts/stm32mp151.dtsi | 2 +-
arch/arm/boot/dts/sun8i-h2-plus-orangepi-zero.dts | 2 +-
arch/arm/include/asm/domain.h | 13 --
arch/arm/include/asm/mach/map.h | 1 +
arch/arm/include/asm/ptrace.h | 26 +++
arch/arm/mm/Kconfig | 6 +-
arch/arm/mm/alignment.c | 3 +
arch/arm/mm/mmu.c | 15 +-
arch/arm/mm/proc-v7-bugs.c | 9 +-
arch/arm/probes/decode.h | 26 +--
arch/arm64/boot/dts/broadcom/bcm4908/bcm4906.dtsi | 8 +
arch/arm64/boot/dts/broadcom/bcm4908/bcm4908.dtsi | 2 +
arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi | 5 +-
arch/powerpc/sysdev/xive/spapr.c | 5 +-
arch/riscv/boot/dts/microchip/microchip-mpfs.dtsi | 4 +
arch/riscv/kvm/vcpu.c | 2 +
arch/s390/Makefile | 8 +-
arch/s390/lib/Makefile | 3 +-
arch/s390/lib/expoline/Makefile | 3 +
arch/s390/lib/{ => expoline}/expoline.S | 0
arch/sh/include/asm/io.h | 8 +-
arch/x86/include/asm/cpufeatures.h | 1 +
arch/x86/include/asm/msr-index.h | 3 +
arch/x86/include/asm/nospec-branch.h | 6 +
arch/x86/include/asm/setup.h | 3 +
arch/x86/kernel/acpi/cppc.c | 6 +
arch/x86/kernel/cpu/amd.c | 23 +-
arch/x86/kernel/cpu/bugs.c | 13 ++
arch/x86/kernel/cpu/common.c | 19 ++
arch/x86/kernel/cpu/cpu.h | 2 +
arch/x86/kernel/cpu/hygon.c | 6 +
arch/x86/kernel/head64.c | 4 +-
arch/x86/kvm/emulate.c | 28 +--
arch/x86/kvm/x86.c | 18 +-
arch/x86/mm/init.c | 14 +-
arch/x86/mm/mem_encrypt_boot.S | 10 +-
arch/x86/xen/enlighten_pv.c | 8 +-
arch/x86/xen/setup.c | 6 +-
arch/x86/xen/xen-asm.S | 20 +-
arch/x86/xen/xen-head.S | 11 +-
arch/x86/xen/xen-ops.h | 6 +-
drivers/acpi/acpi_video.c | 11 +-
drivers/base/cpu.c | 8 +
drivers/cpufreq/pmac32-cpufreq.c | 4 +
drivers/firmware/sysfb.c | 58 ++++-
drivers/firmware/sysfb_simplefb.c | 16 +-
drivers/gpio/gpio-sim.c | 16 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 25 ++-
drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c | 3 +-
drivers/gpu/drm/amd/amdgpu/dce_v10_0.c | 3 +-
drivers/gpu/drm/amd/amdgpu/dce_v11_0.c | 3 +-
drivers/gpu/drm/amd/amdgpu/dce_v6_0.c | 3 +-
drivers/gpu/drm/amd/amdgpu/dce_v8_0.c | 3 +-
drivers/gpu/drm/amd/amdkfd/kfd_device.c | 2 +
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 85 ++++++-
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 8 +
.../amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 17 ++
drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 11 +-
drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c | 2 +
drivers/gpu/drm/i915/display/intel_dp_mst.c | 1 +
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 11 +-
drivers/gpu/drm/i915/gt/intel_gt.c | 15 +-
drivers/gpu/drm/i915/gt/intel_reset.c | 37 ++-
drivers/gpu/drm/i915/gt/selftest_lrc.c | 8 +-
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 9 +
drivers/gpu/drm/i915/gvt/cmd_parser.c | 6 +-
drivers/gpu/drm/i915/i915_scatterlist.c | 19 +-
drivers/gpu/drm/i915/i915_scatterlist.h | 6 +-
drivers/gpu/drm/i915/intel_region_ttm.c | 10 +-
drivers/gpu/drm/i915/intel_region_ttm.h | 3 +-
drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 2 +-
.../gpu/drm/i915/selftests/intel_memory_region.c | 21 +-
drivers/gpu/drm/i915/selftests/mock_region.c | 3 +-
drivers/gpu/drm/panfrost/panfrost_drv.c | 4 +-
drivers/gpu/drm/panfrost/panfrost_mmu.c | 2 +-
drivers/irqchip/irq-or1k-pic.c | 1 -
drivers/net/can/xilinx_can.c | 4 +-
.../net/ethernet/aquantia/atlantic/aq_pci_func.c | 23 +-
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 5 +-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 +
drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 8 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 13 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 7 +-
.../chelsio/inline_crypto/chtls/chtls_cm.c | 2 +-
drivers/net/ethernet/faraday/ftgmac100.c | 15 +-
drivers/net/ethernet/intel/ice/ice_devids.h | 1 +
drivers/net/ethernet/intel/ice/ice_devlink.c | 59 +++--
drivers/net/ethernet/intel/ice/ice_fw_update.c | 96 +++++++-
drivers/net/ethernet/intel/ice/ice_main.c | 1 +
.../ethernet/marvell/prestera/prestera_router.c | 1 +
drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 20 +-
.../ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c | 3 +-
.../ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c | 3 +-
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 39 +++-
.../net/ethernet/mellanox/mlx5/core/esw/legacy.c | 5 +-
drivers/net/ethernet/mscc/ocelot_fdma.c | 17 +-
drivers/net/ethernet/netronome/nfp/nfdk/dp.c | 33 ++-
drivers/net/ethernet/sfc/ef10.c | 3 +
drivers/net/ethernet/sfc/ef10_sriov.c | 10 +-
.../ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c | 1 +
.../net/ethernet/stmicro/stmmac/dwmac-ingenic.c | 6 +-
drivers/net/ethernet/ti/am65-cpsw-nuss.c | 17 +-
drivers/net/phy/sfp.c | 2 +-
drivers/net/xen-netback/rx.c | 1 +
drivers/nfc/nxp-nci/i2c.c | 8 +-
drivers/nvme/host/core.c | 8 +-
drivers/nvme/host/nvme.h | 1 +
drivers/nvme/host/pci.c | 3 +-
drivers/nvme/host/rdma.c | 12 +-
drivers/nvme/host/tcp.c | 13 +-
drivers/nvme/host/trace.h | 2 +-
drivers/pinctrl/aspeed/pinctrl-aspeed.c | 4 +-
drivers/pinctrl/freescale/pinctrl-imx93.c | 1 +
drivers/platform/x86/hp-wmi.c | 3 +
drivers/platform/x86/intel/pmc/core.c | 1 +
drivers/platform/x86/thinkpad_acpi.c | 50 ++---
drivers/power/supply/power_supply_core.c | 24 +-
drivers/s390/crypto/ap_bus.c | 2 +-
drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 7 +
drivers/scsi/megaraid/megaraid_sas_base.c | 3 +
drivers/scsi/ufs/ufshcd.c | 2 +-
drivers/soc/ixp4xx/ixp4xx-npe.c | 2 +-
drivers/spi/spi-amd.c | 8 +
drivers/tee/tee_core.c | 2 +-
drivers/tty/pty.c | 14 +-
drivers/tty/serial/8250/8250_core.c | 4 +
drivers/tty/serial/8250/8250_port.c | 4 +-
drivers/tty/serial/amba-pl011.c | 23 +-
drivers/tty/serial/mvebu-uart.c | 25 ++-
drivers/tty/serial/samsung_tty.c | 5 +-
drivers/tty/serial/serial_core.c | 5 -
drivers/tty/serial/stm32-usart.c | 2 +
drivers/tty/tty.h | 3 +
drivers/tty/tty_buffer.c | 46 +++-
drivers/tty/vt/vt.c | 2 +-
drivers/usb/dwc3/gadget.c | 4 +-
drivers/usb/serial/ftdi_sio.c | 3 +
drivers/usb/serial/ftdi_sio_ids.h | 6 +
drivers/usb/typec/class.c | 1 +
drivers/vdpa/mlx5/net/mlx5_vnet.c | 31 ++-
drivers/vdpa/vdpa_user/vduse_dev.c | 60 +++--
drivers/video/fbdev/core/fbmem.c | 12 +
drivers/virtio/virtio_mmio.c | 26 +++
drivers/xen/gntdev.c | 6 +-
fs/afs/file.c | 2 +-
fs/btrfs/inode.c | 14 +-
fs/btrfs/zoned.c | 13 +-
fs/ceph/addr.c | 11 +-
fs/cifs/smb2pdu.c | 13 +-
fs/exec.c | 2 +-
fs/ksmbd/transport_tcp.c | 2 +-
fs/lockd/svcsubs.c | 14 +-
fs/netfs/buffered_read.c | 17 +-
fs/nfsd/nfs4xdr.c | 9 +
fs/nfsd/nfsd.h | 3 +-
fs/nilfs2/nilfs.h | 3 +
fs/remap_range.c | 3 +-
include/linux/cgroup-defs.h | 3 +-
include/linux/cpu.h | 2 +
include/linux/kexec.h | 6 +
include/linux/netfs.h | 2 +-
include/linux/nvme.h | 2 +
include/linux/reset.h | 2 +-
include/linux/sched/task.h | 2 +-
include/linux/serial_core.h | 5 +
include/linux/sysfb.h | 22 +-
include/net/netfilter/nf_conntrack.h | 9 +-
include/net/netfilter/nf_conntrack_ecache.h | 4 +-
include/net/netfilter/nf_tables.h | 30 ++-
include/net/netns/conntrack.h | 1 -
include/net/raw.h | 2 +-
include/net/sock.h | 2 +-
include/net/tls.h | 4 +-
include/trace/events/sock.h | 6 +-
kernel/cgroup/cgroup.c | 37 +--
kernel/exit.c | 2 +-
kernel/kexec_file.c | 11 +-
kernel/signal.c | 8 +-
kernel/sysctl.c | 57 ++---
kernel/time/posix-timers.c | 19 +-
kernel/trace/trace.c | 11 +-
kernel/trace/trace_events_hist.c | 2 +
mm/damon/vaddr.c | 3 +-
mm/memory.c | 27 +--
mm/sparse-vmemmap.c | 8 +
mm/userfaultfd.c | 5 +-
net/8021q/vlan_netlink.c | 10 +-
net/bridge/br_netfilter_hooks.c | 21 +-
net/core/filter.c | 1 -
net/ipv4/af_inet.c | 4 +-
net/ipv4/cipso_ipv4.c | 12 +-
net/ipv4/fib_semantics.c | 4 +-
net/ipv4/fib_trie.c | 2 +-
net/ipv4/icmp.c | 20 +-
net/ipv4/inet_timewait_sock.c | 3 +-
net/ipv4/inetpeer.c | 12 +-
net/ipv4/nexthop.c | 5 +-
net/ipv4/syncookies.c | 2 +-
net/ipv4/sysctl_net_ipv4.c | 12 +
net/ipv4/tcp.c | 3 +-
net/ipv4/tcp_input.c | 2 +-
net/ipv4/tcp_output.c | 4 +-
net/ipv6/icmp.c | 2 +-
net/ipv6/route.c | 2 +-
net/ipv6/seg6_iptunnel.c | 5 +-
net/ipv6/seg6_local.c | 2 -
net/mac80211/wme.c | 4 +-
net/mptcp/protocol.c | 4 +-
net/netfilter/nf_conntrack_core.c | 86 ++++---
net/netfilter/nf_conntrack_ecache.c | 139 ++++++------
net/netfilter/nf_conntrack_netlink.c | 125 ++++++++---
net/netfilter/nf_conntrack_standalone.c | 3 +
net/netfilter/nf_log_syslog.c | 8 +-
net/netfilter/nf_tables_api.c | 72 ++++--
net/netfilter/nf_tables_core.c | 24 +-
net/netfilter/nf_tables_trace.c | 44 ++--
net/tipc/socket.c | 1 +
net/tls/tls_device.c | 4 +-
net/tls/tls_main.c | 7 +-
security/integrity/evm/evm_crypto.c | 7 +-
security/integrity/ima/ima_appraise.c | 3 +-
security/integrity/ima/ima_crypto.c | 1 +
security/integrity/ima/ima_efi.c | 2 +
sound/pci/hda/patch_conexant.c | 1 +
sound/pci/hda/patch_realtek.c | 20 ++
sound/soc/codecs/cs35l41-lib.c | 10 +-
sound/soc/codecs/cs35l41.c | 12 +-
sound/soc/codecs/cs47l15.c | 5 +-
sound/soc/codecs/madera.c | 14 +-
sound/soc/codecs/max98373-sdw.c | 12 +-
sound/soc/codecs/rt1308-sdw.c | 11 +
sound/soc/codecs/rt1316-sdw.c | 11 +
sound/soc/codecs/rt5682-sdw.c | 5 +-
sound/soc/codecs/rt700-sdw.c | 6 +-
sound/soc/codecs/rt700.c | 14 +-
sound/soc/codecs/rt711-sdca-sdw.c | 9 +-
sound/soc/codecs/rt711-sdca.c | 18 +-
sound/soc/codecs/rt711-sdw.c | 9 +-
sound/soc/codecs/rt711.c | 16 +-
sound/soc/codecs/rt715-sdca-sdw.c | 12 +
sound/soc/codecs/rt715-sdw.c | 12 +
sound/soc/codecs/sgtl5000.c | 9 +
sound/soc/codecs/sgtl5000.h | 1 +
sound/soc/codecs/tas2764.c | 46 ++--
sound/soc/codecs/tas2764.h | 6 +-
sound/soc/codecs/wcd9335.c | 8 +-
sound/soc/codecs/wcd938x.c | 12 +
sound/soc/codecs/wm5110.c | 8 +-
sound/soc/codecs/wm_adsp.c | 2 +-
sound/soc/intel/boards/bytcr_wm5102.c | 13 +-
sound/soc/intel/boards/sof_sdw.c | 51 +++--
sound/soc/intel/skylake/skl-nhlt.c | 40 ++--
sound/soc/soc-dapm.c | 5 +
sound/soc/soc-ops.c | 4 +-
sound/soc/sof/intel/hda-dsp.c | 10 +-
sound/soc/sof/intel/hda-loader.c | 10 +-
sound/soc/sof/intel/hda.h | 1 +
sound/usb/quirks-table.h | 248 +++++++++++++++++++++
sound/usb/quirks.c | 9 +
tools/objtool/check.c | 19 +-
tools/testing/selftests/wireguard/qemu/Makefile | 5 +-
.../selftests/wireguard/qemu/arch/arm.config | 1 +
.../selftests/wireguard/qemu/arch/armeb.config | 1 +
.../selftests/wireguard/qemu/arch/i686.config | 1 +
.../selftests/wireguard/qemu/arch/m68k.config | 1 +
.../selftests/wireguard/qemu/arch/mips.config | 1 +
.../selftests/wireguard/qemu/arch/mipsel.config | 1 +
.../selftests/wireguard/qemu/arch/powerpc.config | 1 +
tools/testing/selftests/wireguard/qemu/init.c | 11 +
277 files changed, 2493 insertions(+), 955 deletions(-)Greg Kroah-Hartman
maintainerPATCH 1/231 USB: serial: ftdi_sio: add Belimo device ids
From: Lucien Buchmann <[email protected]>
commit 7c239a071d1f04b7137789810807b4108d475c72 upstream.
Those two product ids are known.
--- drivers/usb/serial/ftdi_sio.c | 3 +++ drivers/usb/serial/ftdi_sio_ids.h | 6 ++++++ 2 files changed, 9 insertions(+)
drivers/usb/serial/ftdi_sio.c +3 -0
@@ -1023,6 +1023,9 @@ static const struct usb_device_id id_tab { USB_DEVICE(FTDI_VID, CHETCO_SEASMART_DISPLAY_PID) }, { USB_DEVICE(FTDI_VID, CHETCO_SEASMART_LITE_PID) }, { USB_DEVICE(FTDI_VID, CHETCO_SEASMART_ANALOG_PID) }, +/* Belimo Automation devices */ + { USB_DEVICE(FTDI_VID, BELIMO_ZTH_PID) }, + { USB_DEVICE(FTDI_VID, BELIMO_ZIP_PID) },/* ICP DAS I-756xU devices */ { USB_DEVICE(ICPDAS_VID, ICPDAS_I7560U_PID) }, { USB_DEVICE(ICPDAS_VID, ICPDAS_I7561U_PID) },
drivers/usb/serial/ftdi_sio_ids.h +6 -0
@@ -1569,6 +1569,12 @@#define CHETCO_SEASMART_ANALOG_PID 0xA5AF/* SeaSmart Analog Adapter */ /* +* Belimo Automation +*/ +#define BELIMO_ZTH_PID 0x8050 +#define BELIMO_ZIP_PID 0xC811 + +/* * Unjo AB */ #define UNJO_VID 0x22B7
Greg Kroah-Hartman
maintainerPATCH 2/231 usb: typec: add missing uevent when partner support PD
In reply to Greg Kroah-HartmanFrom: Linyu Yuan <[email protected]>
commit 6fb9e1d94789e8ee5a258a23bc588693f743fd6c upstream.
System like Android allow user control power role from UI, it is possible to implement application base on typec uevent to refresh UI, but found there is chance that UI show different state from typec attribute file.
In typec_set_pwr_opmode(), when partner support PD, there is no uevent send to user space which cause the problem.
Fix it by sending uevent notification when change power mode to PD.
--- drivers/usb/typec/class.c | 1 + 1 file changed, 1 insertion(+)
drivers/usb/typec/class.c +1 -0
@@ -1718,6 +1718,7 @@ void typec_set_pwr_opmode(struct typec_p partner->usb_pd =1 ;sysfs_notify (&partner_dev->kobj,NULL ,"supports_usb_power_delivery" ); +kobject_uevent (&partner_dev->kobj,KOBJ_CHANGE ); }put_device (partner_dev); }
Greg Kroah-Hartman
maintainerPATCH 3/231 usb: dwc3: gadget: Fix event pending check
In reply to Greg Kroah-HartmanFrom: Thinh Nguyen <[email protected]>
commit 7441b273388b9a59d8387a03ffbbca9d5af6348c upstream.
The DWC3_EVENT_PENDING flag is used to protect against invalid call to top-half interrupt handler, which can occur when there's a delay in software detection of the interrupt line deassertion.
However, the clearing of this flag was done prior to unmasking the interrupt line, creating opportunity where the top-half handler can come. This breaks the serialization and creates a race between the top-half and bottom-half handler, resulting in losing synchronization between the controller and the driver when processing events.
To fix this, make sure the clearing of the DWC3_EVENT_PENDING is done at the end of the bottom-half handler.
--- drivers/usb/dwc3/gadget.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/usb/dwc3/gadget.c +3 -1
@@ -4224,7 +4224,6 @@ static irqreturn_t dwc3_process_event_bu } evt->count =0 ; - evt->flags &= ~DWC3_EVENT_PENDING ; ret =IRQ_HANDLED ;/* Unmask interrupt */
@@ -4236,6 +4235,9 @@ static irqreturn_t dwc3_process_event_budwc3_writel (dwc->regs,DWC3_DEV_IMOD (0 ), dwc->imod_interval); } +/* Keep the clearing of DWC3_EVENT_PENDING at the end */ + evt->flags &= ~DWC3_EVENT_PENDING ; +return ret; }
Greg Kroah-Hartman
maintainerPATCH 4/231 gpio: sim: fix the chip_name configfs item
In reply to Greg Kroah-HartmanFrom: Bartosz Golaszewski <[email protected]>
commit 7329b071729645e243b6207e76bca2f4951c991b upstream.
The chip_name configs attribute always displays the device name of the first GPIO bank because the logic of the relevant function is simply wrong.
Fix it by correctly comparing the bank's swnode against the GPIO device's children.
--- drivers/gpio/gpio-sim.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-)
drivers/gpio/gpio-sim.c +5 -11
@@ -991,28 +991,22 @@ static struct configfs_attribute *gpio_sim_device_config_attrs[] = { };struct gpio_sim_chip_name_ctx { -struct gpio_sim_device *dev; +struct fwnode_handle *swnode;char *page; };static int gpio_sim_emit_chip_name (struct device *dev,void *data) {struct gpio_sim_chip_name_ctx *ctx = data; -struct fwnode_handle *swnode; -struct gpio_sim_bank *bank;/* This would be the sysfs device exported in /sys/class/gpio. */ if (dev->class)return 0 ; - swnode =dev_fwnode (dev); +if (device_match_fwnode (dev, ctx->swnode)) +return sprintf (ctx->page,"%s\n" ,dev_name (dev)); -list_for_each_entry (bank, &ctx->dev->bank_list, siblings) { -if (bank->swnode == swnode) -return sprintf (ctx->page,"%s\n" ,dev_name (dev)); - } - -return -ENODATA ; +return 0 ; } static ssize_t gpio_sim_bank_config_chip_name_show(struct config_item *item,
@@ -1020,7 +1014,7 @@ static ssize_t gpio_sim_bank_config_chip_name_show(struct config_item *item, {struct gpio_sim_bank *bank =to_gpio_sim_bank (item);struct gpio_sim_device *dev =gpio_sim_bank_get_device (bank); -struct gpio_sim_chip_name_ctx ctx = { dev, page }; +struct gpio_sim_chip_name_ctx ctx = { bank->swnode, page };int ret;mutex_lock (&dev->lock);
Signature · 1 line
2.37.1
Greg Kroah-Hartman
maintainerPATCH 5/231 tty: serial: samsung_tty: set dma burst_size to 1
In reply to Greg Kroah-HartmanFrom: Chanho Park <[email protected]>
commit f7e35e4bf1e8dc2c8cbd5e0955dc1bd58558dae0 upstream.
The src_maxburst and dst_maxburst have been changed to 1 but the settings of the UCON register aren't changed yet. They should be changed as well according to the dmaengine slave config.
--- drivers/tty/serial/samsung_tty.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/tty/serial/samsung_tty.c +2 -3
@@ -377,8 +377,7 @@ static void enable_tx_dma(struct s3c24xx/* Enable tx dma mode */ ucon =rd_regl (port,S3C2410_UCON ); ucon &= ~(S3C64XX_UCON_TXBURST_MASK |S3C64XX_UCON_TXMODE_MASK ); - ucon |= (dma_get_cache_alignment () >=16 ) ? -S3C64XX_UCON_TXBURST_16 :S3C64XX_UCON_TXBURST_1 ; + ucon |=S3C64XX_UCON_TXBURST_1 ; ucon |=S3C64XX_UCON_TXMODE_DMA ;wr_regl (port,S3C2410_UCON , ucon);
@@ -674,7 +673,7 @@ static void enable_rx_dma(struct s3c24xxS3C64XX_UCON_DMASUS_EN |S3C64XX_UCON_TIMEOUT_EN | S3C64XX_UCON_RXMODE_MASK); - ucon |=S3C64XX_UCON_RXBURST_16 | + ucon |=S3C64XX_UCON_RXBURST_1 |0xf <<S3C64XX_UCON_TIMEOUT_SHIFT |S3C64XX_UCON_EMPTYINT_EN | S3C64XX_UCON_TIMEOUT_EN |
Greg Kroah-Hartman
maintainerPATCH 6/231 x86/xen: Use clear_bss() for Xen PV guests
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 96e8fc5818686d4a1591bb6907e7fdb64ef29884 upstream.
Instead of clearing the bss area in assembly code, use the clear_bss() function.
This requires to pass the start_info address as parameter to xen_start_kernel() in order to avoid the xen_start_info being zeroed again.
--- arch/x86/include/asm/setup.h | 3 +++ arch/x86/kernel/head64.c | 2 +- arch/x86/xen/enlighten_pv.c | 8 ++++++-- arch/x86/xen/xen-head.S | 10 +--------- 4 files changed, 11 insertions(+), 12 deletions(-)
arch/x86/include/asm/setup.h +3 -0
@@ -132,6 +132,9 @@ void *extend_brk(size_t size, size_t ali }extern void probe_roms (void ); + +void clear_bss (void ); +#ifdef __i386__ asmlinkage void __init i386_start_kernel(void);
arch/x86/kernel/head64.c +1 -1
@@ -421,7 +421,7 @@ void __init do_early_exception(struct pt/* Don't add a printk in there. printk relies on the PDA which is not initialized yet. */ -static void __init clear_bss (void ) +void __init clear_bss (void ) {memset (__bss_start,0 , (unsigned long ) __bss_stop - (unsigned long ) __bss_start);
arch/x86/xen/enlighten_pv.c +6 -2
@@ -1183,15 +1183,19 @@ static void __init xen_domu_set_legacy_fextern void early_xen_iret_patch (void );/* First C function to be called on Xen boot */ -asmlinkage __visible void __init xen_start_kernel(void) +asmlinkage __visible void __init xen_start_kernel(struct start_info *si) {struct physdev_set_iopl set_iopl;unsigned long initrd_start =0 ;int rc; -if (!xen_start_info) +if (!si)return ; +clear_bss (); + + xen_start_info = si; +__text_gen_insn (&early_xen_iret_patch,JMP32_INSN_OPCODE , &early_xen_iret_patch, &xen_iret,JMP32_INSN_SIZE );
arch/x86/xen/xen-head.S +1 -9
@@ -48,15 +48,6 @@ SYM_CODE_START(startup_xen) ANNOTATE_NOENDBR cld - /* Clear .bss */ - xor %eax,%eax - mov $__bss_start, %rdi - mov $__bss_stop, %rcx - sub %rdi, %rcx - shr $3, %rcx - rep stosq - - mov %rsi, xen_start_info mov initial_stack(%rip), %rsp /* Set up %gs.
@@ -71,6 +62,7 @@ SYM_CODE_START(startup_xen) cdq wrmsr + mov %rsi, %rdi call xen_start_kernel SYM_CODE_END(startup_xen) __FINIT
Greg Kroah-Hartman
maintainerPATCH 7/231 ALSA: hda - Add fixup for Dell Latitidue E5430
In reply to Greg Kroah-HartmanFrom: Meng Tang <[email protected]>
commit 841bdf85c226803a78a9319af9b2caa9bf3e2eda upstream.
Another Dell model, another fixup entry: Latitude E5430 needs the same fixup as other Latitude E series as workaround for noise problems.
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -8947,6 +8947,7 @@ static const struct snd_pci_quirk alc269SND_PCI_QUIRK (0x1025 ,0x1430 ,"Acer TravelMate B311R-31" ,ALC256_FIXUP_ACER_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x1025 ,0x1466 ,"Acer Aspire A515-56" ,ALC255_FIXUP_ACER_HEADPHONE_AND_MIC ),SND_PCI_QUIRK (0x1028 ,0x0470 ,"Dell M101z" ,ALC269_FIXUP_DELL_M101Z ), +SND_PCI_QUIRK (0x1028 ,0x053c ,"Dell Latitude E5430" ,ALC292_FIXUP_DELL_E7X ),SND_PCI_QUIRK (0x1028 ,0x054b ,"Dell XPS one 2710" ,ALC275_FIXUP_DELL_XPS ),SND_PCI_QUIRK (0x1028 ,0x05bd ,"Dell Latitude E6440" ,ALC292_FIXUP_DELL_E7X ), SND_PCI_QUIRK(0x1028, 0x05be, "Dell Latitude E6540", ALC292_FIXUP_DELL_E7X),
Greg Kroah-Hartman
maintainerPATCH 8/231 ALSA: hda/conexant: Apply quirk for another HP ProDesk 600 G3 model
In reply to Greg Kroah-HartmanFrom: Meng Tang <[email protected]>
commit d16d69bf5a25d91c6d8f3e29711be12551bf56cd upstream.
There is another HP ProDesk 600 G3 model with the PCI SSID 103c:82b4 that requires the quirk HP_MIC_NO_PRESENCE. Add the corresponding entry to the quirk table.
--- sound/pci/hda/patch_conexant.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_conexant.c +1 -0
@@ -944,6 +944,7 @@ static const struct snd_pci_quirk cxt506SND_PCI_QUIRK (0x103c ,0x828c ,"HP EliteBook 840 G4" ,CXT_FIXUP_HP_DOCK ),SND_PCI_QUIRK (0x103c ,0x8299 ,"HP 800 G3 SFF" ,CXT_FIXUP_HP_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x103c ,0x829a ,"HP 800 G3 DM" ,CXT_FIXUP_HP_MIC_NO_PRESENCE ), +SND_PCI_QUIRK (0x103c ,0x82b4 ,"HP ProDesk 600 G3" ,CXT_FIXUP_HP_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x103c ,0x836e ,"HP ProBook 455 G5" ,CXT_FIXUP_MUTE_LED_GPIO ),SND_PCI_QUIRK (0x103c ,0x837f ,"HP ProBook 470 G5" ,CXT_FIXUP_MUTE_LED_GPIO ), SND_PCI_QUIRK(0x103c, 0x83b2, "HP EliteBook 840 G5", CXT_FIXUP_HP_DOCK),
Greg Kroah-Hartman
maintainerPATCH 9/231 ALSA: hda/realtek: Fix headset mic for Acer SF313-51
In reply to Greg Kroah-HartmanFrom: Meng Tang <[email protected]>
commit 5f3fe25e70559fa3b096ab17e13316c93ddb7020 upstream.
The issue on Acer SWIFT SF313-51 is that headset microphone doesn't work. The following quirk fixed headset microphone issue. Note that the fixup of SF314-54/55 (ALC256_FIXUP_ACER_HEADSET_MIC) was not successful on my SF313-51.
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -8938,6 +8938,7 @@ static const struct snd_pci_quirk alc269SND_PCI_QUIRK (0x1025 ,0x1290 ,"Acer Veriton Z4860G" ,ALC286_FIXUP_ACER_AIO_HEADSET_MIC ),SND_PCI_QUIRK (0x1025 ,0x1291 ,"Acer Veriton Z4660G" ,ALC286_FIXUP_ACER_AIO_HEADSET_MIC ),SND_PCI_QUIRK (0x1025 ,0x129c ,"Acer SWIFT SF314-55" ,ALC256_FIXUP_ACER_HEADSET_MIC ), +SND_PCI_QUIRK (0x1025 ,0x129d ,"Acer SWIFT SF313-51" ,ALC256_FIXUP_ACER_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x1025 ,0x1300 ,"Acer SWIFT SF314-56" ,ALC256_FIXUP_ACER_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x1025 ,0x1308 ,"Acer Aspire Z24-890" ,ALC286_FIXUP_ACER_AIO_HEADSET_MIC ), SND_PCI_QUIRK(0x1025, 0x132a, "Acer TravelMate B114-21", ALC233_FIXUP_ACER_HEADSET_MIC),
Greg Kroah-Hartman
maintainerPATCH 10/231 ALSA: hda/realtek - Fix headset mic problem for a HP machine with alc671
In reply to Greg Kroah-HartmanFrom: Meng Tang <[email protected]>
commit dbe75d314748e08fc6e4576d153d8a69621ee5ca upstream.
On a HP 288 Pro G6, the front mic could not be detected.In order to get it working, the pin configuration needs to be set correctly, and the ALC671_FIXUP_HP_HEADSET_MIC2 fixup needs to be applied.
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -11271,6 +11271,7 @@ static const struct snd_pci_quirk alc662SND_PCI_QUIRK (0x103c ,0x1632 ,"HP RP5800" ,ALC662_FIXUP_HP_RP5800 ),SND_PCI_QUIRK (0x103c ,0x8719 ,"HP" ,ALC897_FIXUP_HP_HSMIC_VERB ),SND_PCI_QUIRK (0x103c ,0x873e ,"HP" ,ALC671_FIXUP_HP_HEADSET_MIC2 ), +SND_PCI_QUIRK (0x103c ,0x877e ,"HP 288 Pro G6" ,ALC671_FIXUP_HP_HEADSET_MIC2 ),SND_PCI_QUIRK (0x103c ,0x885f ,"HP 288 Pro G8" ,ALC671_FIXUP_HP_HEADSET_MIC2 ),SND_PCI_QUIRK (0x1043 ,0x1080 ,"Asus UX501VW" ,ALC668_FIXUP_HEADSET_MODE ), SND_PCI_QUIRK(0x1043, 0x11cd, "Asus N550", ALC662_FIXUP_ASUS_Nx50),
Greg Kroah-Hartman
maintainerPATCH 11/231 ALSA: hda/realtek: fix mute/micmute LEDs for HP machines
In reply to Greg Kroah-HartmanFrom: Jeremy Szu <[email protected]>
commit 61d307855eb1a2ae849da445edd5389db8a58a5c upstream.
The HP ProBook 440/450 G9 and EliteBook 640/650 G9 have multiple motherboard design and they are using different subsystem ID of audio codec. Add the same quirk for other MBs.
--- sound/pci/hda/patch_realtek.c | 4 ++++ 1 file changed, 4 insertions(+)
sound/pci/hda/patch_realtek.c +4 -0
@@ -9150,6 +9150,10 @@ static const struct snd_pci_quirk alc269SND_PCI_QUIRK (0x103c ,0x89c6 ,"Zbook Fury 17 G9" ,ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED ),SND_PCI_QUIRK (0x103c ,0x89ca ,"HP" ,ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF ),SND_PCI_QUIRK (0x103c ,0x8a78 ,"HP Dev One" ,ALC285_FIXUP_HP_LIMIT_INT_MIC_BOOST ), +SND_PCI_QUIRK (0x103c ,0x8aa0 ,"HP ProBook 440 G9 (MB 8A9E)" ,ALC236_FIXUP_HP_GPIO_LED ), +SND_PCI_QUIRK (0x103c ,0x8aa3 ,"HP ProBook 450 G9 (MB 8AA1)" ,ALC236_FIXUP_HP_GPIO_LED ), +SND_PCI_QUIRK (0x103c ,0x8aa8 ,"HP EliteBook 640 G9 (MB 8AA6)" ,ALC236_FIXUP_HP_GPIO_LED ), +SND_PCI_QUIRK (0x103c ,0x8aab ,"HP EliteBook 650 G9 (MB 8AA9)" ,ALC236_FIXUP_HP_GPIO_LED ),SND_PCI_QUIRK (0x1043 ,0x103e ,"ASUS X540SA" ,ALC256_FIXUP_ASUS_MIC ),SND_PCI_QUIRK (0x1043 ,0x103f ,"ASUS TX300" ,ALC282_FIXUP_ASUS_TX300 ), SND_PCI_QUIRK(0x1043, 0x106d, "Asus K53BE", ALC269_FIXUP_LIMIT_INT_MIC_BOOST),
Greg Kroah-Hartman
maintainerPATCH 12/231 ALSA: hda/realtek - Fix headset mic problem for a HP machine with alc221
In reply to Greg Kroah-HartmanFrom: Meng Tang <[email protected]>
commit 4ba5c853d7945b3855c3dcb293f7f9f019db641e upstream.
On a HP 288 Pro G2 MT (X9W02AV), the front mic could not be detected. In order to get it working, the pin configuration needs to be set correctly, and the ALC221_FIXUP_HP_288PRO_MIC_NO_PRESENCE fixup needs to be applied.
--- sound/pci/hda/patch_realtek.c | 12 ++++++++++++ 1 file changed, 12 insertions(+)
sound/pci/hda/patch_realtek.c +12 -0
@@ -6953,6 +6953,7 @@ enum {ALC298_FIXUP_LENOVO_SPK_VOLUME ,ALC256_FIXUP_DELL_INSPIRON_7559_SUBWOOFER ,ALC269_FIXUP_ATIV_BOOK_8 , +ALC221_FIXUP_HP_288PRO_MIC_NO_PRESENCE ,ALC221_FIXUP_HP_MIC_NO_PRESENCE ,ALC256_FIXUP_ASUS_HEADSET_MODE , ALC256_FIXUP_ASUS_MIC,
@@ -7889,6 +7890,16 @@ static const struct hda_fixup alc269_fix .chained = true, .chain_id =ALC269_FIXUP_NO_SHUTUP }, + [ALC221_FIXUP_HP_288PRO_MIC_NO_PRESENCE] = { + .type = HDA_FIXUP_PINS, + .v.pins = (const struct hda_pintbl []) { + {0x19 ,0x01a1913c },/* use as headset mic, without its own jack detect */ + {0x1a ,0x01813030 },/* use as headphone mic, without its own jack detect */ + { } + }, + .chained = true, + .chain_id =ALC269_FIXUP_HEADSET_MODE + }, [ALC221_FIXUP_HP_MIC_NO_PRESENCE] = { .type = HDA_FIXUP_PINS, .v.pins = (const struct hda_pintbl []) {
@@ -9064,6 +9075,7 @@ static const struct snd_pci_quirk alc269SND_PCI_QUIRK (0x103c ,0x2335 ,"HP" ,ALC269_FIXUP_HP_MUTE_LED_MIC1 ),SND_PCI_QUIRK (0x103c ,0x2336 ,"HP" ,ALC269_FIXUP_HP_MUTE_LED_MIC1 ),SND_PCI_QUIRK (0x103c ,0x2337 ,"HP" ,ALC269_FIXUP_HP_MUTE_LED_MIC1 ), +SND_PCI_QUIRK (0x103c ,0x2b5e ,"HP 288 Pro G2 MT" ,ALC221_FIXUP_HP_288PRO_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x103c ,0x802e ,"HP Z240 SFF" ,ALC221_FIXUP_HP_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x103c ,0x802f ,"HP Z240" ,ALC221_FIXUP_HP_MIC_NO_PRESENCE ), SND_PCI_QUIRK(0x103c, 0x8077, "HP", ALC256_FIXUP_HP_HEADSET_MIC),
Greg Kroah-Hartman
maintainerPATCH 13/231 ALSA: hda/realtek - Enable the headset-mic on a Xiaomis laptop
In reply to Greg Kroah-HartmanFrom: Meng Tang <[email protected]>
commit 9b043a8f386485c74c0f8eea2c287d5bdbdf3279 upstream.
The headset on this machine is not defined, after applying the quirk ALC256_FIXUP_ASUS_HEADSET_MIC, the headset-mic works well
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -9425,6 +9425,7 @@ static const struct snd_pci_quirk alc269SND_PCI_QUIRK (0x1d72 ,0x1602 ,"RedmiBook" ,ALC255_FIXUP_XIAOMI_HEADSET_MIC ),SND_PCI_QUIRK (0x1d72 ,0x1701 ,"XiaomiNotebook Pro" ,ALC298_FIXUP_DELL1_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x1d72 ,0x1901 ,"RedmiBook 14" ,ALC256_FIXUP_ASUS_HEADSET_MIC ), +SND_PCI_QUIRK (0x1d72 ,0x1945 ,"Redmi G" ,ALC256_FIXUP_ASUS_HEADSET_MIC ),SND_PCI_QUIRK (0x1d72 ,0x1947 ,"RedmiBook Air" ,ALC255_FIXUP_XIAOMI_HEADSET_MIC ),SND_PCI_QUIRK (0x8086 ,0x2074 ,"Intel NUC 8" ,ALC233_FIXUP_INTEL_NUC8_DMIC ), SND_PCI_QUIRK(0x8086, 0x2080, "Intel NUC 8 Rugged", ALC256_FIXUP_INTEL_NUC8_RUGGED),
Greg Kroah-Hartman
maintainerPATCH 14/231 xen/netback: avoid entering xenvif_rx_next_skb() with an empty rx queue
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 94e8100678889ab428e68acadf042de723f094b9 upstream.
xenvif_rx_next_skb() is expecting the rx queue not being empty, but in case the loop in xenvif_rx_action() is doing multiple iterations, the availability of another skb in the rx queue is not being checked.
This can lead to crashes:
[40072.537261] BUG: unable to handle kernel NULL pointer dereference at 0000000000000080 [40072.537407] IP: xenvif_rx_skb+0x23/0x590 [xen_netback] [40072.537534] PGD 0 P4D 0 [40072.537644] Oops: 0000 [#1] SMP NOPTI [40072.537749] CPU: 0 PID: 12505 Comm: v1-c40247-q2-gu Not tainted 4.12.14-122.121-default #1 SLE12-SP5 [40072.537867] Hardware name: HP ProLiant DL580 Gen9/ProLiant DL580 Gen9, BIOS U17 11/23/2021 [40072.537999] task: ffff880433b38100 task.stack: ffffc90043d40000 [40072.538112] RIP: e030:xenvif_rx_skb+0x23/0x590 [xen_netback] [40072.538217] RSP: e02b:ffffc90043d43de0 EFLAGS: 00010246 [40072.538319] RAX: 0000000000000000 RBX: ffffc90043cd7cd0 RCX: 00000000000000f7 [40072.538430] RDX: 0000000000000000 RSI: 0000000000000006 RDI: ffffc90043d43df8 [40072.538531] RBP: 000000000000003f R08: 000077ff80000000 R09: 0000000000000008 [40072.538644] R10: 0000000000007ff0 R11: 00000000000008f6 R12: ffffc90043ce2708 [40072.538745] R13: 0000000000000000 R14: ffffc90043d43ed0 R15: ffff88043ea748c0 [40072.538861] FS: 0000000000000000(0000) GS:ffff880484600000(0000) knlGS:0000000000000000 [40072.538988] CS: e033 DS: 0000 ES: 0000 CR0: 0000000080050033 [40072.539088] CR2: 0000000000000080 CR3: 0000000407ac8000 CR4: 0000000000040660 [40072.539211] Call Trace: [40072.539319] xenvif_rx_action+0x71/0x90 [xen_netback] [40072.539429] xenvif_kthread_guest_rx+0x14a/0x29c [xen_netback]
Fix that by stopping the loop in case the rx queue becomes empty.
--- drivers/net/xen-netback/rx.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/xen-netback/rx.c +1 -0
@@ -495,6 +495,7 @@ void xenvif_rx_action(struct xenvif_queu queue->rx_copy.completed = &completed_skbs;while (xenvif_rx_ring_slots_available (queue) && + !skb_queue_empty (&queue->rx_queue) && work_done <RX_BATCH_SIZE ) {xenvif_rx_skb (queue); work_done++;
Greg Kroah-Hartman
maintainerPATCH 15/231 fix race between exit_itimers() and /proc/pid/timers
In reply to Greg Kroah-HartmanFrom: Oleg Nesterov <[email protected]>
commit d5b36a4dbd06c5e8e36ca8ccc552f679069e2946 upstream.
As Chris explains, the comment above exit_itimers() is not correct, we can race with proc_timers_seq_ops. Change exit_itimers() to clear signal->posix_timers with ->siglock held.
--- fs/exec.c | 2 +- include/linux/sched/task.h | 2 +- kernel/exit.c | 2 +- kernel/time/posix-timers.c | 19 ++++++++++++++----- 4 files changed, 17 insertions(+), 8 deletions(-)
fs/exec.c +1 -1
@@ -1297,7 +1297,7 @@ int begin_new_exec(struct linux_binprm * bprm->mm =NULL ;#ifdef CONFIG_POSIX_TIMERS -exit_itimers (me->signal); +exit_itimers (me);flush_itimer_signals ();#endif
include/linux/sched/task.h +1 -1
@@ -83,7 +83,7 @@ static inline void exit_thread(struct taextern __noreturn void do_group_exit (int );extern void exit_files (struct task_struct *); -extern void exit_itimers (struct signal_struct *); +extern void exit_itimers (struct task_struct *);extern pid_t kernel_clone (struct kernel_clone_args *kargs);struct task_struct *create_io_thread (int (*fn)(void *),void *arg,int node);
kernel/exit.c +1 -1
@@ -766,7 +766,7 @@ void __noreturn do_exit(long code)#ifdef CONFIG_POSIX_TIMERS hrtimer_cancel (&tsk->signal->real_timer); -exit_itimers (tsk->signal); +exit_itimers (tsk);#endif if (tsk->mm)setmax_mm_hiwater_rss (&tsk->signal->maxrss, tsk->mm);
kernel/time/posix-timers.c +14 -5
@@ -1051,15 +1051,24 @@ retry_delete: }/* -* This is called by do_exit or de_thread, only when there are no more -* references to the shared signal_struct. +* This is called by do_exit or de_thread, only when nobody else can +* modify the signal->posix_timers list. Yet we need sighand->siglock +* to prevent the race with /proc/pid/timers. */ -void exit_itimers (struct signal_struct *sig) +void exit_itimers (struct task_struct *tsk) { +struct list_head timers;struct k_itimer *tmr; -while (!list_empty (&sig->posix_timers)) { - tmr = list_entry(sig->posix_timers.next, struct k_itimer, list); +if (list_empty (&tsk->signal->posix_timers)) +return ; + +spin_lock_irq (&tsk->sighand->siglock); +list_replace_init (&tsk->signal->posix_timers, &timers); +spin_unlock_irq (&tsk->sighand->siglock); + +while (!list_empty (&timers)) { + tmr = list_first_entry(&timers, struct k_itimer, list);itimer_delete (tmr); } }
Greg Kroah-Hartman
maintainerPATCH 16/231 mm: userfaultfd: fix UFFDIO_CONTINUE on fallocated shmem pages
In reply to Greg Kroah-HartmanFrom: Axel Rasmussen <[email protected]>
commit 73f37dbcfe1763ee2294c7717a1f571e27d17fd8 upstream.
When fallocate() is used on a shmem file, the pages we allocate can end up with !PageUptodate.
Since UFFDIO_CONTINUE tries to find the existing page the user wants to map with SGP_READ, we would fail to find such a page, since shmem_getpage_gfp returns with a "NULL" pagep for SGP_READ if it discovers !PageUptodate. As a result, UFFDIO_CONTINUE returns -EFAULT, as it would do if the page wasn't found in the page cache at all.
This isn't the intended behavior. UFFDIO_CONTINUE is just trying to find if a page exists, and doesn't care whether it still needs to be cleared or not. So, instead of SGP_READ, pass in SGP_NOALLOC. This is the same, except for one critical difference: in the !PageUptodate case, SGP_NOALLOC will clear the page and then return it. With this change, UFFDIO_CONTINUE works properly (succeeds) on a shmem file which has been fallocated, but otherwise not modified.
--- mm/userfaultfd.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
mm/userfaultfd.c +4 -1
@@ -231,7 +231,10 @@ static int mcontinue_atomic_pte(struct mstruct page *page;int ret; - ret =shmem_getpage (inode, pgoff, &page,SGP_READ ); + ret =shmem_getpage (inode, pgoff, &page,SGP_NOALLOC ); +/* Our caller expects us to return -EFAULT if we failed to find page. */ +if (ret == -ENOENT ) + ret = -EFAULT ;if (ret) goto out;if (!page) {
Greg Kroah-Hartman
maintainerPATCH 17/231 mm: sparsemem: fix missing higher order allocation splitting
In reply to Greg Kroah-HartmanFrom: Muchun Song <[email protected]>
commit 39d35edee4537487e5178f258e23518272a66413 upstream.
Higher order allocations for vmemmap pages from buddy allocator must be able to be treated as indepdenent small pages as they can be freed individually by the caller. There is no problem for higher order vmemmap pages allocated at boot time since each individual small page will be initialized at boot time. However, it will be an issue for memory hotplug case since those higher order vmemmap pages are allocated from buddy allocator without initializing each individual small page's refcount. The system will panic in put_page_testzero() when CONFIG_DEBUG_VM is enabled if the vmemmap page is freed.
--- mm/sparse-vmemmap.c | 8 ++++++++ 1 file changed, 8 insertions(+)
mm/sparse-vmemmap.c +8 -0
@@ -78,6 +78,14 @@ static int __split_vmemmap_huge_pmd(pmd_t *pmd, unsigned long start)spin_lock (&init_mm.page_table_lock);if (likely (pmd_leaf (*pmd))) { +/* +* Higher order allocations from buddy allocator must be able to +* be treated as indepdenent small pages (as they can be freed +* individually). +*/ +if (!PageReserved (page)) +split_page (page,get_order (PMD_SIZE )); +/* Make pte visible before pmd. See comment in pmd_install(). */ smp_wmb ();pmd_populate_kernel (&init_mm, pmd, pgtable);
Signature · 1 line
2.37.1
Greg Kroah-Hartman
maintainerPATCH 18/231 mm: split huge PUD on wp_huge_pud fallback
In reply to Greg Kroah-HartmanFrom: Gowans, James <[email protected]>
commit 14c99d65941538aa33edd8dc7b1bbbb593c324a2 upstream.
Currently the implementation will split the PUD when a fallback is taken inside the create_huge_pud function. This isn't where it should be done:
the splitting should be done in wp_huge_pud, just like it's done for PMDs. Reason being that if a callback is taken during create, there is no PUD yet so nothing to split, whereas if a fallback is taken when encountering a write protection fault there is something to split.
It looks like this was the original intention with the commit where the splitting was introduced, but somehow it got moved to the wrong place between v1 and v2 of the patch series. Rebase mistake perhaps.
--- mm/memory.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-)
mm/memory.c +14 -13
@@ -4528,6 +4528,19 @@ static vm_fault_t create_huge_pud(struct defined(CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD) /* No support for anonymous transparent PUD pages yet */ if (vma_is_anonymous(vmf->vma)) + return VM_FAULT_FALLBACK; +if (vmf->vma->vm_ops->huge_fault) +return vmf->vma->vm_ops->huge_fault (vmf,PE_SIZE_PUD ); +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ +return VM_FAULT_FALLBACK ; +} + +static vm_fault_t wp_huge_pud (struct vm_fault *vmf,pud_t orig_pud) +{ +#if defined(CONFIG_TRANSPARENT_HUGEPAGE ) && \ + defined(CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD ) +/* No support for anonymous transparent PUD pages yet */ +if (vma_is_anonymous (vmf->vma)) goto split;if (vmf->vma->vm_ops->huge_fault) {vm_fault_t ret = vmf->vma->vm_ops->huge_fault (vmf,PE_SIZE_PUD );
@@ -4538,19 +4551,7 @@ static vm_fault_t create_huge_pud(struct split:/* COW or write-notify not handled on PUD level: split pud.*/ __split_huge_pud (vmf->vma, vmf->pud, vmf->address); -#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ -return VM_FAULT_FALLBACK ; -} - -static vm_fault_t wp_huge_pud (struct vm_fault *vmf,pud_t orig_pud) -{ -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -/* No support for anonymous transparent PUD pages yet */ -if (vma_is_anonymous (vmf->vma)) -return VM_FAULT_FALLBACK ; -if (vmf->vma->vm_ops->huge_fault) -return vmf->vma->vm_ops->huge_fault (vmf,PE_SIZE_PUD ); -#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ +#endif /* CONFIG_TRANSPARENT_HUGEPAGE && CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD */ return VM_FAULT_FALLBACK ; }
Greg Kroah-Hartman
maintainerPATCH 19/231 mm/damon: use set_huge_pte_at() to make huge pte old
In reply to Greg Kroah-HartmanFrom: Baolin Wang <[email protected]>
commit ed1523a895ffdabcab6e067af18685ed00f5ce15 upstream.
The huge_ptep_set_access_flags() can not make the huge pte old according to the discussion [1], that means we will always mornitor the young state of the hugetlb though we stopped accessing the hugetlb, as a result DAMON will get inaccurate accessing statistics.
So changing to use set_huge_pte_at() to make the huge pte old to fix this issue.
[1] https://lore.kernel.org/all/[email protected]/
--- mm/damon/vaddr.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
mm/damon/vaddr.c +1 -2
@@ -407,8 +407,7 @@ static void damon_hugetlb_mkold(pte_t *pif (pte_young (entry)) { referenced = true; entry =pte_mkold (entry); -huge_ptep_set_access_flags (vma, addr, pte, entry, - vma->vm_flags &VM_WRITE ); +set_huge_pte_at (mm, addr, pte, entry); }#ifdef CONFIG_MMU_NOTIFIER
Greg Kroah-Hartman
maintainerPATCH 20/231 tracing/histograms: Fix memory leak problem
In reply to Greg Kroah-HartmanFrom: Zheng Yejian <[email protected]>
commit 7edc3945bdce9c39198a10d6129377a5c53559c2 upstream.
This reverts commit 46bbe5c671e06f070428b9be142cc4ee5cedebac.
As commit 46bbe5c671e0 ("tracing: fix double free") said, the "double free" problem reported by clang static analyzer is:
Quoted 4 lines
In
parse_var_defs()if there is a problem allocating var_defs.expr, the earlier var_defs.name is freed. This free is duplicated byfree_var_defs()which frees the rest of the list.
However, if there is a problem allocating N-th var_defs.expr:
+ in parse_var_defs(), the freed 'earlier var_defs.name' is
actually the N-th var_defs.name;
+ then in free_var_defs(), the names from 0th to (N-1)-th are freed;
IF ALLOCATING PROBLEM HAPPENED HERE!!! -+
\
|
0th 1th (N-1)-th N-th V
+-------------+-------------+-----+-------------+-----------
var_defs: | name | expr | name | expr | ... | name | expr | name | ///
+-------------+-------------+-----+-------------+-----------These two frees don't act on same name, so there was no "double free" problem before. Conversely, after that commit, we get a "memory leak" problem because the above "N-th var_defs.name" is not freed.
If enable CONFIG_DEBUG_KMEMLEAK and inject a fault at where the N-th var_defs.expr allocated, then execute on shell like:
$ echo 'hist:key=call_site:val=$v1,$v2:v1=bytes_req,v2=bytes_alloc' > \ /sys/kernel/debug/tracing/events/kmem/kmalloc/trigger
Then kmemleak reports:
unreferenced object 0xffff8fb100ef3518 (size 8):
comm "bash", pid 196, jiffies 4295681690 (age 28.538s)
hex dump (first 8 bytes):
76 31 00 00 b1 8f ff ff v1......
backtrace:
[<0000000038fe4895>] kstrdup+0x2d/0x60
[<00000000c99c049a>] event_hist_trigger_parse+0x206f/0x20e0
[<00000000ae70d2cc>] trigger_process_regex+0xc0/0x110
[<0000000066737a4c>] event_trigger_write+0x75/0xd0
[<000000007341e40c>] vfs_write+0xbb/0x2a0
[<0000000087fde4c2>] ksys_write+0x59/0xd0
[<00000000581e9cdf>] do_syscall_64+0x3a/0x80
[<00000000cf3b065c>] entry_SYSCALL_64_after_hwframe+0x46/0xb0--- kernel/trace/trace_events_hist.c | 2 ++ 1 file changed, 2 insertions(+)
kernel/trace/trace_events_hist.c +2 -0
@@ -4429,6 +4429,8 @@ static int parse_var_defs(struct hist_tr s =kstrdup (field_str,GFP_KERNEL );if (!s) { +kfree (hist_data->attrs->var_defs.name[n_vars]); + hist_data->attrs->var_defs.name[n_vars] =NULL ; ret = -ENOMEM ; goto free; }
Greg Kroah-Hartman
maintainerPATCH 21/231 net: sock: tracing: Fix sock_exceed_buf_limit not to dereference stale pointer
In reply to Greg Kroah-HartmanFrom: Steven Rostedt (Google) <[email protected]>
commit 820b8963adaea34a87abbecb906d1f54c0aabfb7 upstream.
The trace event sock_exceed_buf_limit saves the prot->sysctl_mem pointer and then dereferences it in the TP_printk() portion. This is unsafe as the TP_printk() portion is executed at the time the buffer is read. That is, it can be seconds, minutes, days, months, even years later. If the proto is freed, then this dereference will can also lead to a kernel crash.
Instead, save the sysctl_mem array into the ring buffer and have the TP_printk() reference that instead. This is the proper and safe way to read pointers in trace events.
--- include/trace/events/sock.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
include/trace/events/sock.h +4 -2
@@ -98,7 +98,7 @@ TRACE_EVENT(sock_exceed_buf_limit,TP_STRUCT__entry ( __array(char, name, 32) - __field(long *, sysctl_mem) + __array(long, sysctl_mem, 3) __field(long, allocated) __field(int, sysctl_rmem) __field(int, rmem_alloc)
@@ -110,7 +110,9 @@ TRACE_EVENT(sock_exceed_buf_limit,TP_fast_assign ( strncpy(__entry->name, prot->name, 32); - __entry->sysctl_mem = prot->sysctl_mem; + __entry->sysctl_mem[0 ] =READ_ONCE (prot->sysctl_mem[0 ]); + __entry->sysctl_mem[1 ] =READ_ONCE (prot->sysctl_mem[1 ]); + __entry->sysctl_mem[2 ] =READ_ONCE (prot->sysctl_mem[2 ]); __entry->allocated = allocated; __entry->sysctl_rmem =sk_get_rmem0 (sk, prot); __entry->rmem_alloc =atomic_read (&sk->sk_rmem_alloc);
Greg Kroah-Hartman
maintainerPATCH 22/231 ip: fix dflt addr selection for connected nexthop
In reply to Greg Kroah-HartmanFrom: Nicolas Dichtel <[email protected]>
commit 747c14307214b55dbd8250e1ab44cad8305756f1 upstream.
When a nexthop is added, without a gw address, the default scope was set to 'host'. Thus, when a source address is selected, 127.0.0.1 may be chosen but rejected when the route is used.
When using a route without a nexthop id, the scope can be configured in the route, thus the problem doesn't exist.
To explain more deeply: when a user creates a nexthop, it cannot specify the scope. To create it, the function nh_create_ipv4() calls fib_check_nh() with scope set to 0. fib_check_nh() calls fib_check_nh_nongw() wich was setting scope to 'host'. Then, nh_create_ipv4() calls fib_info_update_nhc_saddr() with scope set to 'host'. The src addr is chosen before the route is inserted.
When a 'standard' route (ie without a reference to a nexthop) is added, fib_create_info() calls fib_info_update_nhc_saddr() with the scope set by the user. iproute2 set the scope to 'link' by default.
Here is a way to reproduce the problem:
ip netns add foo
ip -n foo link set lo up
ip netns add bar
ip -n bar link set lo up
sleep 1
ip -n foo link add name eth0 type dummy
ip -n foo link set eth0 up
ip -n foo address add 192.168.0.1/24 dev eth0
ip -n foo link add name veth0 type veth peer name veth1 netns bar ip -n foo link set veth0 up
ip -n bar link set veth1 up
ip -n bar address add 192.168.1.1/32 dev veth1
ip -n bar route add default dev veth1
ip -n foo nexthop add id 1 dev veth0
ip -n foo route add 192.168.1.1 nhid 1
Try to get/use the route:
Quoted 4 lines
$ ip -n foo route get 192.168.1.1 RTNETLINK answers: Invalid argument $ ip netns exec foo ping -c1 192.168.1.1 ping: connect: Invalid argument
Try without nexthop group (iproute2 sets scope to 'link' by dflt):
ip -n foo route del 192.168.1.1
ip -n foo route add 192.168.1.1 dev veth0
Try to get/use the route:
Quoted 10 lines
$ ip -n foo route get 192.168.1.1 192.168.1.1 dev veth0 src 192.168.0.1 uid 0 cache $ ip netns exec foo ping -c1 192.168.1.1 PING 192.168.1.1 (192.168.1.1) 56(84) bytes of data. 64 bytes from 192.168.1.1: icmp_seq=1 ttl=64 time=0.039 ms--- 192.168.1.1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.039/0.039/0.039/0.000 ms
--- net/ipv4/fib_semantics.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/fib_semantics.c +1 -1
@@ -1230,7 +1230,7 @@ static int fib_check_nh_nongw(struct net nh->fib_nh_dev = in_dev->dev;dev_hold_track (nh->fib_nh_dev, &nh->fib_nh_dev_tracker,GFP_ATOMIC ); - nh->fib_nh_scope =RT_SCOPE_HOST ; + nh->fib_nh_scope =RT_SCOPE_LINK ;if (!netif_carrier_ok (nh->fib_nh_dev)) nh->fib_nh_flags |=RTNH_F_LINKDOWN ; err =0 ;
Greg Kroah-Hartman
maintainerPATCH 23/231 ARM: 9213/1: Print message about disabled Spectre workarounds only once
In reply to Greg Kroah-HartmanFrom: Dmitry Osipenko <[email protected]>
commit e4ced82deb5fb17222fb82e092c3f8311955b585 upstream.
Print the message about disabled Spectre workarounds only once. The message is printed each time CPU goes out from idling state on NVIDIA Tegra boards, causing storm in KMSG that makes system unusable.
--- arch/arm/mm/proc-v7-bugs.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
arch/arm/mm/proc-v7-bugs.c +1 -2
@@ -108,8 +108,7 @@ static unsigned int spectre_v2_install_w#else static unsigned int spectre_v2_install_workaround (unsigned int method) { -pr_info ("CPU%u: Spectre V2: workarounds disabled by configuration\n" , -smp_processor_id ()); +pr_info_once ("Spectre V2: workarounds disabled by configuration\n" );return SPECTRE_VULNERABLE ; }
Greg Kroah-Hartman
maintainerPATCH 24/231 ARM: 9214/1: alignment: advance IT state after emulating Thumb instruction
In reply to Greg Kroah-HartmanFrom: Ard Biesheuvel <[email protected]>
commit e5c46fde75e43c15a29b40e5fc5641727f97ae47 upstream.
After emulating a misaligned load or store issued in Thumb mode, we have to advance the IT state by hand, or it will get out of sync with the actual instruction stream, which means we'll end up applying the wrong condition code to subsequent instructions. This might corrupt the program state rather catastrophically.
So borrow the it_advance() helper from the probing code, and use it on CPSR if the emulated instruction is Thumb.
--- arch/arm/include/asm/ptrace.h | 26 ++++++++++++++++++++++++++ arch/arm/mm/alignment.c | 3 +++ arch/arm/probes/decode.h | 26 +------------------------- 3 files changed, 30 insertions(+), 25 deletions(-)
arch/arm/include/asm/ptrace.h +26 -0
@@ -163,5 +163,31 @@ static inline unsigned long user_stack_p ((current_stack_pointer | (THREAD_SIZE -1 )) -7 ) -1 ; \ }) + +/* +* Update ITSTATE after normal execution of an IT block instruction. +* +* The 8 IT state bits are split into two parts in CPSR: +* ITSTATE<1:0> are in CPSR<26:25> +* ITSTATE<7:2> are in CPSR<15:10> +*/ +static inline unsigned long it_advance (unsigned long cpsr) +{ +if ((cpsr &0x06000400 ) ==0 ) { +/* ITSTATE<2:0> == 0 means end of IT block, so clear IT state */ + cpsr &= ~PSR_IT_MASK ; + }else { +/* We need to shift left ITSTATE<4:0> */ +const unsigned long mask =0x06001c00 ;/* Mask ITSTATE<4:0> */ +unsigned long it = cpsr & mask; + it <<=1 ; + it |= it >> (27 -10 );/* Carry ITSTATE<2> to correct place */ + it &= mask; + cpsr &= ~mask; + cpsr |= it; + } +return cpsr; +} +#endif /* __ASSEMBLY__ */ #endif
arch/arm/mm/alignment.c +3 -0
@@ -935,6 +935,9 @@ do_alignment(unsigned long addr, unsigneif (type ==TYPE_LDST )do_alignment_finish_ldst (addr, instr, regs, offset); +if (thumb_mode (regs)) + regs->ARM_cpsr =it_advance (regs->ARM_cpsr); +return 0 ; bad_or_fault:
arch/arm/probes/decode.h +1 -25
@@ -14,6 +14,7 @@#include <linux/types.h> #include <linux/stddef.h> #include <asm/probes.h> +#include <asm/ptrace.h> #include <asm/kprobes.h> void __init arm_probes_decode_init (void );
@@ -35,31 +36,6 @@ void __init find_str_pc_offset(void);#endif -/* -* Update ITSTATE after normal execution of an IT block instruction. -* -* The 8 IT state bits are split into two parts in CPSR: -* ITSTATE<1:0> are in CPSR<26:25> -* ITSTATE<7:2> are in CPSR<15:10> -*/ -static inline unsigned long it_advance (unsigned long cpsr) - { -if ((cpsr &0x06000400 ) ==0 ) { -/* ITSTATE<2:0> == 0 means end of IT block, so clear IT state */ - cpsr &= ~PSR_IT_MASK ; - }else { -/* We need to shift left ITSTATE<4:0> */ -const unsigned long mask =0x06001c00 ;/* Mask ITSTATE<4:0> */ -unsigned long it = cpsr & mask; - it <<=1 ; - it |= it >> (27 -10 );/* Carry ITSTATE<2> to correct place */ - it &= mask; - cpsr &= ~mask; - cpsr |= it; - } -return cpsr; -} -static inline void __kprobes bx_write_pc (long pcv,struct pt_regs *regs) {long cpsr = regs->ARM_cpsr;
Greg Kroah-Hartman
maintainerPATCH 25/231 wifi: mac80211: fix queue selection for mesh/OCB interfaces
In reply to Greg Kroah-HartmanFrom: Felix Fietkau <[email protected]>
commit 50e2ab39291947b6c6c7025cf01707c270fcde59 upstream.
When using iTXQ, the code assumes that there is only one vif queue for broadcast packets, using the BE queue. Allowing non-BE queue marking violates that assumption and txq->ac == skb_queue_mapping is no longer guaranteed. This can cause issues with queue handling in the driver and also causes issues with the recent ATF change, resulting in an AQL underflow warning.
--- net/mac80211/wme.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/mac80211/wme.c +2 -2
@@ -147,8 +147,8 @@ u16 __ieee80211_select_queue(struct ieeebool qos;/* all mesh/ocb stations are required to support WME */ -if (sdata->vif.type ==NL80211_IFTYPE_MESH_POINT || - sdata->vif.type ==NL80211_IFTYPE_OCB ) +if (sta && (sdata->vif.type ==NL80211_IFTYPE_MESH_POINT || + sdata->vif.type ==NL80211_IFTYPE_OCB )) qos = true;else if (sta) qos = sta->sta.wme;
Greg Kroah-Hartman
maintainerPATCH 26/231 cgroup: Use separate src/dst nodes when preloading css_sets for migration
In reply to Greg Kroah-HartmanFrom: Tejun Heo <[email protected]>
commit 07fd5b6cdf3cc30bfde8fe0f644771688be04447 upstream.
Each cset (css_set) is pinned by its tasks. When we're moving tasks around across csets for a migration, we need to hold the source and destination csets to ensure that they don't go away while we're moving tasks about. This is done by linking cset->mg_preload_node on either the mgctx->preloaded_src_csets or mgctx->preloaded_dst_csets list. Using the same cset->mg_preload_node for both the src and dst lists was deemed okay as a cset can't be both the source and destination at the same time.
Unfortunately, this overloading becomes problematic when multiple tasks are involved in a migration and some of them are identity noop migrations while others are actually moving across cgroups. For example, this can happen with the following sequence on cgroup1:
#1> mkdir -p /sys/fs/cgroup/misc/a/b
#2> echo $$ > /sys/fs/cgroup/misc/a/cgroup.procs #3> RUN_A_COMMAND_WHICH_CREATES_MULTIPLE_THREADS & #4> PID=$!
#5> echo $PID > /sys/fs/cgroup/misc/a/b/tasks #6> echo $PID > /sys/fs/cgroup/misc/a/cgroup.procs
the process including the group leader back into a. In this final migration, non-leader threads would be doing identity migration while the group leader is doing an actual one.
After #3, let's say the whole process was in cset A, and that after #4, the leader moves to cset B. Then, during #6, the following happens:
1. cgroup_migrate_add_src() is called on B for the leader.
2. cgroup_migrate_add_src() is called on A for the other threads.
3. cgroup_migrate_prepare_dst() is called. It scans the src list.
4. It notices that B wants to migrate to A, so it tries to A to the dst
list but realizes that its ->mg_preload_node is already busy.
5. and then it notices A wants to migrate to A as it's an identity migration, it culls it by list_del_init()'ing its ->mg_preload_node and putting references accordingly.
6. The rest of migration takes place with B on the src list but nothing on the dst list.
This means that A isn't held while migration is in progress. If all tasks leave A before the migration finishes and the incoming task pins it, the cset will be destroyed leading to use-after-free.
This is caused by overloading cset->mg_preload_node for both src and dst preload lists. We wanted to exclude the cset from the src list but ended up inadvertently excluding it from the dst list too.
This patch fixes the issue by separating out cset->mg_preload_node into ->mg_src_preload_node and ->mg_dst_preload_node, so that the src and dst preloadings don't interfere with each other.
--- include/linux/cgroup-defs.h | 3 ++- kernel/cgroup/cgroup.c | 37 +++++++++++++++++++++++-------------- 2 files changed, 25 insertions(+), 15 deletions(-)
include/linux/cgroup-defs.h +2 -1
@@ -264,7 +264,8 @@ struct css_set {* List of csets participating in the on-going migration either as * source or destination. Protected by cgroup_mutex. */ -struct list_head mg_preload_node; +struct list_head mg_src_preload_node; +struct list_head mg_dst_preload_node;struct list_head mg_node;/*
kernel/cgroup/cgroup.c +23 -14
@@ -765,7 +765,8 @@ struct css_set init_css_set = { .task_iters = LIST_HEAD_INIT(init_css_set.task_iters), .threaded_csets = LIST_HEAD_INIT(init_css_set.threaded_csets), .cgrp_links = LIST_HEAD_INIT(init_css_set.cgrp_links), - .mg_preload_node = LIST_HEAD_INIT(init_css_set.mg_preload_node), + .mg_src_preload_node = LIST_HEAD_INIT(init_css_set.mg_src_preload_node), + .mg_dst_preload_node = LIST_HEAD_INIT(init_css_set.mg_dst_preload_node), .mg_node = LIST_HEAD_INIT(init_css_set.mg_node),/*
@@ -1240,7 +1241,8 @@ static struct css_set *find_css_set(struINIT_LIST_HEAD (&cset->threaded_csets);INIT_HLIST_NODE (&cset->hlist);INIT_LIST_HEAD (&cset->cgrp_links); -INIT_LIST_HEAD (&cset->mg_preload_node); +INIT_LIST_HEAD (&cset->mg_src_preload_node); +INIT_LIST_HEAD (&cset->mg_dst_preload_node);INIT_LIST_HEAD (&cset->mg_node);/* Copy the set of subsystem state objects generated in
@@ -2597,21 +2599,27 @@ int cgroup_migrate_vet_dst(struct cgroup*/ void cgroup_migrate_finish (struct cgroup_mgctx *mgctx) { -LIST_HEAD (preloaded);struct css_set *cset, *tmp_cset;lockdep_assert_held (&cgroup_mutex);spin_lock_irq (&css_set_lock); -list_splice_tail_init (&mgctx->preloaded_src_csets, &preloaded); -list_splice_tail_init (&mgctx->preloaded_dst_csets, &preloaded); +list_for_each_entry_safe (cset, tmp_cset, &mgctx->preloaded_src_csets, + mg_src_preload_node) { + cset->mg_src_cgrp =NULL ; + cset->mg_dst_cgrp =NULL ; + cset->mg_dst_cset =NULL ; +list_del_init (&cset->mg_src_preload_node); +put_css_set_locked (cset); + } -list_for_each_entry_safe (cset, tmp_cset, &preloaded, mg_preload_node) { +list_for_each_entry_safe (cset, tmp_cset, &mgctx->preloaded_dst_csets, + mg_dst_preload_node) { cset->mg_src_cgrp =NULL ; cset->mg_dst_cgrp =NULL ; cset->mg_dst_cset =NULL ; -list_del_init (&cset->mg_preload_node); +list_del_init (&cset->mg_dst_preload_node);put_css_set_locked (cset); }
@@ -2651,7 +2659,7 @@ void cgroup_migrate_add_src(struct css_sif (src_cset->dead)return ; -if (!list_empty (&src_cset->mg_preload_node)) +if (!list_empty (&src_cset->mg_src_preload_node))return ; src_cgrp =cset_cgroup_from_root (src_cset, dst_cgrp->root);
@@ -2664,7 +2672,7 @@ void cgroup_migrate_add_src(struct css_s src_cset->mg_src_cgrp = src_cgrp; src_cset->mg_dst_cgrp = dst_cgrp;get_css_set (src_cset); -list_add_tail (&src_cset->mg_preload_node, &mgctx->preloaded_src_csets); +list_add_tail (&src_cset->mg_src_preload_node, &mgctx->preloaded_src_csets); }/**
@@ -2689,7 +2697,7 @@ int cgroup_migrate_prepare_dst(struct cg/* look up the dst cset for each src cset and link it to src */ list_for_each_entry_safe (src_cset, tmp_cset, &mgctx->preloaded_src_csets, - mg_preload_node) { + mg_src_preload_node) {struct css_set *dst_cset;struct cgroup_subsys *ss;int ssid;
@@ -2708,7 +2716,7 @@ int cgroup_migrate_prepare_dst(struct cgif (src_cset == dst_cset) { src_cset->mg_src_cgrp =NULL ; src_cset->mg_dst_cgrp =NULL ; -list_del_init (&src_cset->mg_preload_node); +list_del_init (&src_cset->mg_src_preload_node);put_css_set (src_cset);put_css_set (dst_cset);continue ;
@@ -2716,8 +2724,8 @@ int cgroup_migrate_prepare_dst(struct cg src_cset->mg_dst_cset = dst_cset; -if (list_empty (&dst_cset->mg_preload_node)) -list_add_tail (&dst_cset->mg_preload_node, +if (list_empty (&dst_cset->mg_dst_preload_node)) +list_add_tail (&dst_cset->mg_dst_preload_node, &mgctx->preloaded_dst_csets);else put_css_set (dst_cset);
@@ -2963,7 +2971,8 @@ static int cgroup_update_dfl_csses(struc goto out_finish;spin_lock_irq (&css_set_lock); -list_for_each_entry (src_cset, &mgctx.preloaded_src_csets, mg_preload_node) { +list_for_each_entry (src_cset, &mgctx.preloaded_src_csets, + mg_src_preload_node) {struct task_struct *task, *ntask;/* all tasks in src_csets need to be migrated */
Greg Kroah-Hartman
maintainerPATCH 27/231 btrfs: return -EAGAIN for NOWAIT dio reads/writes on compressed and inline extents
In reply to Greg Kroah-HartmanFrom: Filipe Manana <[email protected]>
commit a4527e1853f8ff6e0b7c2dadad6268bd38427a31 upstream.
When doing a direct IO read or write, we always return -ENOTBLK when we find a compressed extent (or an inline extent) so that we fallback to buffered IO. This however is not ideal in case we are in a NOWAIT context (io_uring for example), because buffered IO can block and we currently have no support for NOWAIT semantics for buffered IO, so if we need to fallback to buffered IO we should first signal the caller that we may need to block by returning -EAGAIN instead.
This behaviour can also result in short reads being returned to user space, which although it's not incorrect and user space should be able to deal with partial reads, it's somewhat surprising and even some popular applications like QEMU (Link tag #1) and MariaDB (Link tag #2) don't deal with short reads properly (or at all).
The short read case happens when we try to read from a range that has a non-compressed and non-inline extent followed by a compressed extent. After having read the first extent, when we find the compressed extent we return -ENOTBLK from btrfs_dio_iomap_begin(), which results in iomap to treat the request as a short read, returning 0 (success) and waiting for previously submitted bios to complete (this happens at fs/iomap/direct-io.c:__iomap_dio_rw()). After that, and while at btrfs_file_read_iter(), we call filemap_read() to use buffered IO to read the remaining data, and pass it the number of bytes we were able to read with direct IO. Than at filemap_read() if we get a page fault error when accessing the read buffer, we return a partial read instead of an -EFAULT error, because the number of bytes previously read is greater than zero.
So fix this by returning -EAGAIN for NOWAIT direct IO when we find a compressed or an inline extent.
--- fs/btrfs/inode.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-)
fs/btrfs/inode.c +13 -1
@@ -7639,7 +7639,19 @@ static int btrfs_dio_iomap_begin(structif (test_bit (EXTENT_FLAG_COMPRESSED , &em->flags) || em->block_start ==EXTENT_MAP_INLINE ) {free_extent_map (em); - ret = -ENOTBLK ; +/* +* If we are in a NOWAIT context, return -EAGAIN in order to +* fallback to buffered IO. This is not only because we can +* block with buffered IO (no support for NOWAIT semantics at +* the moment) but also to avoid returning short reads to user +* space - this happens if we were able to read some data from +* previous non-compressed extents and then when we fallback to +* buffered IO, at btrfs_file_read_iter() by calling +* filemap_read(), we fail to fault in pages for the read buffer, +* in which case filemap_read() returns a short read (the number +* of bytes previously read is > 0, so it does not return -EFAULT). +*/ + ret = (flags &IOMAP_NOWAIT ) ? -EAGAIN : -ENOTBLK ; goto unlock_err; }
Greg Kroah-Hartman
maintainerPATCH 28/231 btrfs: zoned: fix a leaked bioc in read_zone_info
In reply to Greg Kroah-HartmanFrom: Christoph Hellwig <[email protected]>
commit 2963457829decf0c824a443238d251151ed18ff5 upstream.
The bioc would leak on the normal completion path and also on the RAID56 check (but that one won't happen in practice due to the invalid combination with zoned mode).
--- fs/btrfs/zoned.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-)
fs/btrfs/zoned.c +8 -5
@@ -1727,12 +1727,14 @@ static int read_zone_info(struct btrfs_f ret =btrfs_map_sblock (fs_info,BTRFS_MAP_GET_READ_MIRRORS , logical, &mapped_length, &bioc);if (ret || !bioc || mapped_length <PAGE_SIZE ) { -btrfs_put_bioc (bioc); -return -EIO ; + ret = -EIO ; + goto out_put_bioc; } -if (bioc->map_type &BTRFS_BLOCK_GROUP_RAID56_MASK ) -return -EINVAL ; +if (bioc->map_type &BTRFS_BLOCK_GROUP_RAID56_MASK ) { + ret = -EINVAL ; + goto out_put_bioc; + } nofs_flag =memalloc_nofs_save (); nmirrors = (int )bioc->num_stripes;
@@ -1751,7 +1753,8 @@ static int read_zone_info(struct btrfs_fbreak ; }memalloc_nofs_restore (nofs_flag); - +out_put_bioc: +btrfs_put_bioc (bioc);return ret; }
Greg Kroah-Hartman
maintainerPATCH 29/231 drm/panfrost: Put mapping instead of shmem obj on panfrost_mmu_map_fault_addr() error
In reply to Greg Kroah-HartmanFrom: Dmitry Osipenko <[email protected]>
commit fb6e0637ab7ebd8e61fe24f4d663c4bae99cfa62 upstream.
When panfrost_mmu_map_fault_addr() fails, the BO's mapping should be unreferenced and not the shmem object which backs the mapping.
--- drivers/gpu/drm/panfrost/panfrost_mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/panfrost/panfrost_mmu.c +1 -1
@@ -518,7 +518,7 @@ err_map: err_pages:drm_gem_shmem_put_pages (&bo->base); err_bo: -drm_gem_object_put (&bo->base.base); +panfrost_gem_mapping_put (bomapping);return ret; }
Greg Kroah-Hartman
maintainerPATCH 30/231 drm/panfrost: Fix shrinker list corruption by madvise IOCTL
In reply to Greg Kroah-HartmanFrom: Dmitry Osipenko <[email protected]>
commit 9fc33eaaa979d112d10fea729edcd2a2e21aa912 upstream.
Calling madvise IOCTL twice on BO causes memory shrinker list corruption and crashes kernel because BO is already on the list and it's added to the list again, while BO should be removed from the list before it's re-added. Fix it.
--- drivers/gpu/drm/panfrost/panfrost_drv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/panfrost/panfrost_drv.c +2 -2
@@ -432,8 +432,8 @@ static int panfrost_ioctl_madvise(structif (args->retained) {if (args->madv ==PANFROST_MADV_DONTNEED ) -list_add_tail (&bo->base.madv_list, - &pfdev->shrinker_list); +list_move_tail (&bo->base.madv_list, + &pfdev->shrinker_list);else if (args->madv ==PANFROST_MADV_WILLNEED )list_del_init (&bo->base.madv_list); }
Greg Kroah-Hartman
maintainerPATCH 31/231 fs/remap: constrain dedupe of EOF blocks
In reply to Greg Kroah-HartmanFrom: Dave Chinner <[email protected]>
commit 5750676b64a561f7ec920d7c6ba130fc9c7378f3 upstream.
If dedupe of an EOF block is not constrainted to match against only other EOF blocks with the same EOF offset into the block, it can match against any other block that has the same matching initial bytes in it, even if the bytes beyond EOF in the source file do not match.
Fix this by constraining the EOF block matching to only match against other EOF blocks that have identical EOF offsets and data. This allows "whole file dedupe" to continue to work without allowing eof blocks to randomly match against partial full blocks with the same data.
--- fs/remap_range.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
fs/remap_range.c +2 -1
@@ -71,7 +71,8 @@ static int generic_remap_checks(struct f* Otherwise, make sure the count is also block-aligned, having * already confirmed the starting offsets' block alignment. */ -if (pos_in + count == size_in) { +if (pos_in + count == size_in && + (!(remap_flags &REMAP_FILE_DEDUP ) || pos_out + count == size_out)) { bcount =ALIGN (size_in, bs) - pos_in; }else {if (!IS_ALIGNED (count, bs))
Greg Kroah-Hartman
maintainerPATCH 32/231 nilfs2: fix incorrect masking of permission flags for symlinks
In reply to Greg Kroah-HartmanFrom: Ryusuke Konishi <[email protected]>
commit 5924e6ec1585445f251ea92713eb15beb732622a upstream.
The permission flags of newly created symlinks are wrongly dropped on nilfs2 with the current umask value even though symlinks should have 777 (rwxrwxrwx) permissions:
$ umask 0022 $ touch file && ln -s file symlink; ls -l file symlink -rw-r--r--. 1 root root 0 Jun 23 16:29 file lrwxr-xr-x. 1 root root 4 Jun 23 16:29 symlink -> file
This fixes the bug by inserting a missing check that excludes symlinks.
--- fs/nilfs2/nilfs.h | 3 +++ 1 file changed, 3 insertions(+)
fs/nilfs2/nilfs.h +3 -0
@@ -198,6 +198,9 @@ static inline int nilfs_acl_chmod(structstatic inline int nilfs_init_acl (struct inode *inode,struct inode *dir) { +if (S_ISLNK (inode->i_mode)) +return 0 ; + inode->i_mode &= ~current_umask ();return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 33/231 sh: convert nommu io{re,un}map() to static inline functions
In reply to Greg Kroah-HartmanFrom: Geert Uytterhoeven <[email protected]>
commit d684e0a52d36f8939eda30a0f31ee235ee4ee741 upstream.
Recently, nommu iounmap() was converted from a static inline function to a macro again, basically reverting commit 4580ba4ad2e6b8dd ("sh: Convert iounmap() macros to inline functions"). With -Werror, this leads to build failures like:
drivers/iio/adc/xilinx-ams.c: In function `ams_iounmap_ps':
drivers/iio/adc/xilinx-ams.c:1195:14: error: unused variable `ams' [-Werror=unused-variable]
1195 | struct ams *ams = data;
| ^~~
Fix this by replacing the macros for ioremap() and iounmap() by static inline functions, based on <asm-generic/io.h>.
--- arch/sh/include/asm/io.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
arch/sh/include/asm/io.h +6 -2
@@ -271,8 +271,12 @@ static inline void __iomem *ioremap_prot#endif /* CONFIG_HAVE_IOREMAP_PROT */ #else /* CONFIG_MMU */ -#define iounmap (addr) do { } while (0) -#define ioremap (offset, size) ((void __iomem *)(unsigned long)(offset)) +static inline void __iomem *ioremap (phys_addr_t offset,size_t size) +{ +return (void __iomem *)(unsigned long )offset; +} + +static inline void iounmap (volatile void __iomem *addr) { }#endif /* CONFIG_MMU */ #define ioremap_uc ioremap
Greg Kroah-Hartman
maintainerPATCH 34/231 Revert "evm: Fix memleak in init_desc"
In reply to Greg Kroah-HartmanFrom: Xiu Jianfeng <[email protected]>
commit 51dd64bb99e4478fc5280171acd8e1b529eadaf7 upstream.
This reverts commit ccf11dbaa07b328fa469415c362d33459c140a37.
Commit ccf11dbaa07b ("evm: Fix memleak in init_desc") said there is memleak in init_desc. That may be incorrect, as we can see, tmp_tfm is saved in one of the two global variables hmac_tfm or evm_tfm[hash_algo], then if init_desc is called next time, there is no need to alloc tfm again, so in the error path of kmalloc desc or crypto_shash_init(desc), It is not a problem without freeing tmp_tfm.
And also that commit did not reset the global variable to NULL after freeing tmp_tfm and this makes *tfm a dangling pointer which may cause a UAF issue.
--- security/integrity/evm/evm_crypto.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-)
security/integrity/evm/evm_crypto.c +2 -5
@@ -75,7 +75,7 @@ static struct shash_desc *init_desc(char {long rc;const char *algo; -struct crypto_shash **tfm, *tmp_tfm =NULL ; +struct crypto_shash **tfm, *tmp_tfm;struct shash_desc *desc;if (type ==EVM_XATTR_HMAC ) {
@@ -120,16 +120,13 @@ unlock: alloc: desc =kmalloc (sizeof (*desc) +crypto_shash_descsize (*tfm),GFP_KERNEL ); -if (!desc) { -crypto_free_shash (tmp_tfm); +if (!desc)return ERR_PTR (-ENOMEM ); - } desc->tfm = *tfm; rc =crypto_shash_init (desc);if (rc) { -crypto_free_shash (tmp_tfm);kfree (desc);return ERR_PTR (rc); }
Greg Kroah-Hartman
maintainerPATCH 35/231 reset: Fix devm bulk optional exclusive control getter
In reply to Greg Kroah-HartmanFrom: Serge Semin <[email protected]>
[ Upstream commit a57f68ddc8865d59a19783080cc52fb4a11dc209 ]
Most likely due to copy-paste mistake the device managed version of the denoted reset control getter has been implemented with invalid semantic, which can be immediately spotted by having "WARN_ON(shared && acquired)" warning in the system log as soon as the method is called. Anyway let's fix it by altering the boolean arguments passed to the __devm_reset_control_bulk_get() method from
- shared = true, optional = false, acquired = true
to
+ shared = false, optional = true, acquired = true
That's what they were supposed to be in the first place (see the non-devm version of the same method: reset_control_bulk_get_optional_exclusive()).
--- include/linux/reset.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/reset.h +1 -1
@@ -731,7 +731,7 @@ static inline int __must_check devm_reset_control_bulk_get_optional_exclusive(struct device *dev, int num_rstcs, struct reset_control_bulk_data *rstcs) { - return __devm_reset_control_bulk_get(dev, num_rstcs, rstcs, true, false, true); + return __devm_reset_control_bulk_get(dev, num_rstcs, rstcs, false, true, true); }/**
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 36/231 arm64: dts: ls1028a: Update SFP node to include clock
In reply to Greg Kroah-HartmanFrom: Sean Anderson <[email protected]>
[ Upstream commit 3c12e9da3098a30fc82dea01768d355c28e3692d ]
The clocks property is now mandatory. Add it to avoid warning message.
--- arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi +4 -1
@@ -224,9 +224,12 @@ rst: syscon@1e60000 { little-endian; }; - efuse@1e80000 { + sfp: efuse@1e80000 { compatible = "fsl,ls1028a-sfp"; reg = <0x0 0x1e80000 0x0 0x10000>; + clocks = <&clockgen QORIQ_CLK_PLATFORM_PLL + QORIQ_CLK_PLL_DIV(4)>; + clock-names = "sfp"; #address-cells = <1>; #size-cells = <1>;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 37/231 ARM: dts: imx6qdl-ts7970: Fix ngpio typo and count
In reply to Greg Kroah-HartmanFrom: Kris Bahnsen <[email protected]>
[ Upstream commit e95ea0f687e679fcb0a3a67d0755b81ee7d60db0 ]
Device-tree incorrectly used "ngpio" which caused the driver to fallback to 32 ngpios.
This platform has 62 GPIO registers.
--- arch/arm/boot/dts/imx6qdl-ts7970.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/imx6qdl-ts7970.dtsi +1 -1
@@ -226,7 +226,7 @@ gpio8: gpio@28 { reg = <0x28>; #gpio-cells = <2>; gpio-controller; - ngpio = <32>; + ngpios = <62>; }; sgtl5000: codec@a {
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 38/231 riscv: dts: microchip: hook up the mpfs l2cache
In reply to Greg Kroah-HartmanFrom: Conor Dooley <[email protected]>
[ Upstream commit efa310ba00716d7a872bdc5fa1f5545edc9efd69 ]
The initial PolarFire SoC devicetree must have been forked off from the fu540 one prior to the addition of l2cache controller support being added there. When the controller node was added to mpfs.dtsi, it was not hooked up to the CPUs & thus sysfs reports an incorrect cache configuration. Hook it up.
--- arch/riscv/boot/dts/microchip/microchip-mpfs.dtsi | 4 ++++ 1 file changed, 4 insertions(+)
arch/riscv/boot/dts/microchip/microchip-mpfs.dtsi +4 -0
@@ -51,6 +51,7 @@ cpu1: cpu@1 { riscv,isa = "rv64imafdc"; clocks = <&clkcfg CLK_CPU>; tlb-split; + next-level-cache = <&cctrllr>; status = "okay"; cpu1_intc: interrupt-controller {
@@ -78,6 +79,7 @@ cpu2: cpu@2 { riscv,isa = "rv64imafdc"; clocks = <&clkcfg CLK_CPU>; tlb-split; + next-level-cache = <&cctrllr>; status = "okay"; cpu2_intc: interrupt-controller {
@@ -105,6 +107,7 @@ cpu3: cpu@3 { riscv,isa = "rv64imafdc"; clocks = <&clkcfg CLK_CPU>; tlb-split; + next-level-cache = <&cctrllr>; status = "okay"; cpu3_intc: interrupt-controller {
@@ -132,6 +135,7 @@ cpu4: cpu@4 { riscv,isa = "rv64imafdc"; clocks = <&clkcfg CLK_CPU>; tlb-split; + next-level-cache = <&cctrllr>; status = "okay"; cpu4_intc: interrupt-controller { #interrupt-cells = <1>;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 39/231 spi: amd: Limit max transfer and message size
In reply to Greg Kroah-HartmanFrom: Cristian Ciocaltea <[email protected]>
[ Upstream commit 6ece49c56965544262523dae4a071ace3db63507 ]
Enabling the SPI CS35L41 audio codec driver for Steam Deck [1] revealed a problem with the current AMD SPI controller driver implementation, consisting of an unrecoverable system hang.
The issue can be prevented if we ensure the max transfer size and the max message size do not exceed the FIFO buffer size.
According to the implementation of the downstream driver, the AMD SPI controller is not able to handle more than 70 bytes per transfer, which corresponds to the size of the FIFO buffer.
Hence, let's fix this by setting the SPI limits mentioned above.
[1] https://lore.kernel.org/r/[email protected]
--- drivers/spi/spi-amd.c | 8 ++++++++ 1 file changed, 8 insertions(+)
drivers/spi/spi-amd.c +8 -0
@@ -33,6 +33,7 @@#define AMD_SPI_RX_COUNT_REG 0x4B#define AMD_SPI_STATUS_REG 0x4C +#define AMD_SPI_FIFO_SIZE 70#define AMD_SPI_MEM_SIZE 200/* M_CMD OP codes for SPI */
@@ -270,6 +271,11 @@ static int amd_spi_master_transfer(struct spi_master *master,return 0 ; } +static size_t amd_spi_max_transfer_size (struct spi_device *spi) +{ +return AMD_SPI_FIFO_SIZE ; +} +static int amd_spi_probe (struct platform_device *pdev) {struct device *dev = &pdev->dev;
@@ -302,6 +308,8 @@ static int amd_spi_probe(struct platform_device *pdev) master->flags =SPI_MASTER_HALF_DUPLEX ; master->setup = amd_spi_master_setup; master->transfer_one_message = amd_spi_master_transfer; + master->max_transfer_size = amd_spi_max_transfer_size; + master->max_message_size = amd_spi_max_transfer_size;/* Register the controller with SPI framework */ err =devm_spi_register_master (dev, master);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 40/231 ARM: 9209/1: Spectre-BHB: avoid pr_info() every time a CPU comes out of idle
In reply to Greg Kroah-HartmanFrom: Ard Biesheuvel <[email protected]>
[ Upstream commit 0609e200246bfd3b7516091c491bec4308349055 ]
Jon reports that the Spectre-BHB init code is filling up the kernel log with spurious notifications about which mitigation has been enabled, every time any CPU comes out of a low power state.
Given that Spectre-BHB mitigations are system wide, only a single mitigation can be enabled, and we already print an error if two types of CPUs coexist in a single system that require different Spectre-BHB mitigations.
This means that the pr_info() that describes the selected mitigation does not need to be emitted for each CPU anyway, and so we can simply emit it only once.
In order to clarify the above in the log message, update it to describe that the selected mitigation will be enabled on all CPUs, including ones that are unaffected. If another CPU comes up later that is affected and requires a different mitigation, we report an error as before.
--- arch/arm/mm/proc-v7-bugs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
arch/arm/mm/proc-v7-bugs.c +3 -3
@@ -208,10 +208,10 @@ static int spectre_bhb_install_workaround(int method)return SPECTRE_VULNERABLE ; spectre_bhb_method = method; - } -pr_info ("CPU%u: Spectre BHB: using %s workaround\n" , -smp_processor_id (),spectre_bhb_method_name (method)); +pr_info ("CPU%u: Spectre BHB: enabling %s workaround for all CPUs\n" , +smp_processor_id (),spectre_bhb_method_name (method)); + }return SPECTRE_MITIGATED ; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 41/231 ARM: 9210/1: Mark the FDT_FIXED sections as shareable
In reply to Greg Kroah-HartmanFrom: Zhen Lei <[email protected]>
[ Upstream commit 598f0a99fa8a35be44b27106b43ddc66417af3b1 ]
commit 7a1be318f579 ("ARM: 9012/1: move device tree mapping out of linear region") use FDT_FIXED_BASE to map the whole FDT_FIXED_SIZE memory area which contains fdt. But it only reserves the exact physical memory that fdt occupied. Unfortunately, this mapping is non-shareable. An illegal or speculative read access can bring the RAM content from non-fdt zone into cache, PIPT makes it to be hit by subsequently read access through shareable mapping(such as linear mapping), and the cache consistency between cores is lost due to non-shareable property.
|<---------FDT_FIXED_SIZE------>| | |
-------------------------------
| <non-fdt> | <fdt> | <non-fdt> |
-------------------------------
1. CoreA read <non-fdt> through MT_ROM mapping, the old data is loaded into the cache.
2. CoreB write <non-fdt> to update data through linear mapping. CoreA received the notification to invalid the corresponding cachelines, but the property non-shareable makes it to be ignored.
3. CoreA read <non-fdt> through linear mapping, cache hit, the old data is read.
To eliminate this risk, add a new memory type MT_MEMORY_RO. Compared to MT_ROM, it is shareable and non-executable.
Here's an example:
list_del corruption. prev->next should be c0ecbf74, but was c08410dc kernel BUG at lib/list_debug.c:53!
... ...
PC is at __list_del_entry_valid+0x58/0x98
LR is at __list_del_entry_valid+0x58/0x98
psr: 60000093
sp : c0ecbf30 ip : 00000000 fp : 00000001 r10: c08410d0 r9 : 00000001 r8 : c0825e0c r7 : 20000013 r6 : c08410d0 r5 : c0ecbf74 r4 : c0ecbf74 r3 : c0825d08 r2 : 00000000 r1 : df7ce6f4 r0 : 00000044 ... ... Stack: (0xc0ecbf30 to 0xc0ecc000) bf20: c0ecbf74 c0164fd0 c0ecbf70 c0165170 bf40: c0eca000 c0840c00 c0840c00 c0824500 c0825e0c c0189bbc c088f404 60000013 bf60: 60000013 c0e85100 000004ec 00000000 c0ebcdc0 c0ecbf74 c0ecbf74 c0825d08 ... ... < next prev > (__list_del_entry_valid) from (__list_del_entry+0xc/0x20) (__list_del_entry) from (finish_swait+0x60/0x7c) (finish_swait) from (rcu_gp_kthread+0x560/0xa20) (rcu_gp_kthread) from (kthread+0x14c/0x15c) (kthread) from (ret_from_fork+0x14/0x24)
The faulty list node to be deleted is a local variable, its address is c0ecbf74. The dumped stack shows that 'prev' = c0ecbf74, but its value before lib/list_debug.c:53 is c08410dc. A large amount of printing results in swapping out the cacheline containing the old data(MT_ROM mapping is read only, so the cacheline cannot be dirty), and the subsequent dump operation obtains new data from the DDR.
--- arch/arm/include/asm/mach/map.h | 1 + arch/arm/mm/mmu.c | 15 ++++++++++++++- 2 files changed, 15 insertions(+), 1 deletion(-)
arch/arm/include/asm/mach/map.h +1 -0
@@ -27,6 +27,7 @@ enum {MT_HIGH_VECTORS ,MT_MEMORY_RWX ,MT_MEMORY_RW , +MT_MEMORY_RO ,MT_ROM ,MT_MEMORY_RWX_NONCACHED , MT_MEMORY_RW_DTCM,
arch/arm/mm/mmu.c +14 -1
@@ -296,6 +296,13 @@ static struct mem_type mem_types[] __ro_after_init = { .prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE, .domain = DOMAIN_KERNEL, }, + [MT_MEMORY_RO] = { + .prot_pte =L_PTE_PRESENT |L_PTE_YOUNG |L_PTE_DIRTY | + L_PTE_XN | L_PTE_RDONLY, + .prot_l1 = PMD_TYPE_TABLE, + .prot_sect = PMD_TYPE_SECT, + .domain = DOMAIN_KERNEL, + }, [MT_ROM] = { .prot_sect = PMD_TYPE_SECT, .domain = DOMAIN_KERNEL,
@@ -489,6 +496,7 @@ static void __init build_mem_type_table(void)/* Also setup NX memory mapping */ mem_types[MT_MEMORY_RW ].prot_sect |=PMD_SECT_XN ; + mem_types[MT_MEMORY_RO ].prot_sect |=PMD_SECT_XN ; }if (cpu_arch >= CPU_ARCH_ARMv7 && (cr &CR_TRE )) {/*
@@ -568,6 +576,7 @@ static void __init build_mem_type_table(void) mem_types[MT_ROM ].prot_sect |=PMD_SECT_APX |PMD_SECT_AP_WRITE ; mem_types[MT_MINICLEAN ].prot_sect |=PMD_SECT_APX |PMD_SECT_AP_WRITE ; mem_types[MT_CACHECLEAN ].prot_sect |=PMD_SECT_APX |PMD_SECT_AP_WRITE ; + mem_types[MT_MEMORY_RO ].prot_sect |=PMD_SECT_APX |PMD_SECT_AP_WRITE ;#endif /*
@@ -587,6 +596,8 @@ static void __init build_mem_type_table(void) mem_types[MT_MEMORY_RWX ].prot_pte |=L_PTE_SHARED ; mem_types[MT_MEMORY_RW ].prot_sect |=PMD_SECT_S ; mem_types[MT_MEMORY_RW ].prot_pte |=L_PTE_SHARED ; + mem_types[MT_MEMORY_RO ].prot_sect |=PMD_SECT_S ; + mem_types[MT_MEMORY_RO ].prot_pte |=L_PTE_SHARED ; mem_types[MT_MEMORY_DMA_READY ].prot_pte |=L_PTE_SHARED ; mem_types[MT_MEMORY_RWX_NONCACHED ].prot_sect |=PMD_SECT_S ; mem_types[MT_MEMORY_RWX_NONCACHED ].prot_pte |=L_PTE_SHARED ;
@@ -647,6 +658,8 @@ static void __init build_mem_type_table(void) mem_types[MT_MEMORY_RWX ].prot_pte |= kern_pgprot; mem_types[MT_MEMORY_RW ].prot_sect |= ecc_mask | cp->pmd; mem_types[MT_MEMORY_RW ].prot_pte |= kern_pgprot; + mem_types[MT_MEMORY_RO ].prot_sect |= ecc_mask | cp->pmd; + mem_types[MT_MEMORY_RO ].prot_pte |= kern_pgprot; mem_types[MT_MEMORY_DMA_READY ].prot_pte |= kern_pgprot; mem_types[MT_MEMORY_RWX_NONCACHED ].prot_sect |= ecc_mask; mem_types[MT_ROM ].prot_sect |= cp->pmd;
@@ -1360,7 +1373,7 @@ static void __init devicemaps_init(const struct machine_desc *mdesc) map.pfn =__phys_to_pfn (__atags_pointer &SECTION_MASK ); map.virtual =FDT_FIXED_BASE ; map.length =FDT_FIXED_SIZE ; - map.type =MT_ROM ; + map.type =MT_MEMORY_RO ;create_mapping (&map); }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 42/231 net/mlx5e: kTLS, Fix build time constant test in TX
In reply to Greg Kroah-HartmanFrom: Tariq Toukan <[email protected]>
[ Upstream commit 6cc2714e85754a621219693ea8aa3077d6fca0cb ]
Use the correct constant (TLS_DRIVER_STATE_SIZE_TX) in the comparison against the size of the private TX TLS driver context.
--- drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c +1 -2
@@ -68,8 +68,7 @@ mlx5e_set_ktls_tx_priv_ctx(struct tls_context *tls_ctx,struct mlx5e_ktls_offload_context_tx **ctx =__tls_driver_ctx (tls_ctx,TLS_OFFLOAD_CTX_DIR_TX ); -BUILD_BUG_ON (sizeof (struct mlx5e_ktls_offload_context_tx *) > -TLS_OFFLOAD_CONTEXT_SIZE_TX ); +BUILD_BUG_ON (sizeof (priv_tx) >TLS_DRIVER_STATE_SIZE_TX ); *ctx = priv_tx; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 43/231 net/mlx5e: kTLS, Fix build time constant test in RX
In reply to Greg Kroah-HartmanFrom: Tariq Toukan <[email protected]>
[ Upstream commit 2ec6cf9b742a5c18982861322fa5de6510f8f57e ]
Use the correct constant (TLS_DRIVER_STATE_SIZE_RX) in the comparison against the size of the private RX TLS driver context.
--- drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c +1 -2
@@ -231,8 +231,7 @@ mlx5e_set_ktls_rx_priv_ctx(struct tls_context *tls_ctx,struct mlx5e_ktls_offload_context_rx **ctx =__tls_driver_ctx (tls_ctx,TLS_OFFLOAD_CTX_DIR_RX ); -BUILD_BUG_ON (sizeof (struct mlx5e_ktls_offload_context_rx *) > -TLS_OFFLOAD_CONTEXT_SIZE_RX ); +BUILD_BUG_ON (sizeof (priv_rx) >TLS_DRIVER_STATE_SIZE_RX ); *ctx = priv_rx; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 44/231 net/mlx5e: Fix enabling sriov while tc nic rules are offloaded
In reply to Greg Kroah-HartmanFrom: Paul Blakey <[email protected]>
[ Upstream commit 0c9d876545a56aebed30fa306d0460a4d28d271a ]
There is a total of four 4M entries flow tables. In sriov disabled mode, ct, ct_nat and post_act take three of them. When adding the first tc nic rule in this mode, it will take another 4M table for the tc <chain,prio> table. If user then enables sriov, the legacy flow table tries to take another 4M and fails, and so enablement fails.
To fix that, have legacy fdb take the next available maximum size from the fs ft pool.
--- drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c +3 -2
@@ -11,6 +11,7 @@#include "mlx5_core.h" #include "eswitch.h" #include "fs_core.h" +#include "fs_ft_pool.h" #include "esw/qos.h" enum {
@@ -95,8 +96,7 @@ static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw)if (!flow_group_in)return -ENOMEM ; - table_size =BIT (MLX5_CAP_ESW_FLOWTABLE_FDB (dev, log_max_ft_size)); - ft_attr.max_fte = table_size; + ft_attr.max_fte =POOL_NEXT_SIZE ; ft_attr.prio =LEGACY_FDB_PRIO ; fdb =mlx5_create_flow_table (root_ns, &ft_attr);if (IS_ERR (fdb)) {
@@ -105,6 +105,7 @@ static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw) goto out; } esw->fdb_table.legacy.fdb = fdb; + table_size = fdb->max_fte;/* Addresses group : Full match unicast/multicast addresses */ MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 45/231 net/mlx5e: CT: Use own workqueue instead of mlx5e priv
In reply to Greg Kroah-HartmanFrom: Roi Dayan <[email protected]>
[ Upstream commit 6c4e8fa03fde7e5b304594294e397a9ba92feaf6 ]
Allocate a ct priv workqueue instead of using mlx5e priv one so flushing will only be of related CT entries. Also move flushing of the workqueue before rhashtable destroy otherwise entries won't be valid.
--- .../ethernet/mellanox/mlx5/core/en/tc_ct.c | 20 +++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c +12 -8
@@ -66,6 +66,7 @@ struct mlx5_tc_ct_priv {struct mlx5_ct_fs *fs;struct mlx5_ct_fs_ops *fs_ops;spinlock_t ht_lock;/* protects ft entries */ +struct workqueue_struct *wq; };struct mlx5_ct_flow {
@@ -927,14 +928,11 @@ static void mlx5_tc_ct_entry_del_work(struct work_struct *work)static void __mlx5_tc_ct_entry_put (struct mlx5_ct_entry *entry) { -struct mlx5e_priv *priv; -if (!refcount_dec_and_test (&entry->refcnt))return ; - priv =netdev_priv (entry->ct_priv->netdev);INIT_WORK (&entry->work, mlx5_tc_ct_entry_del_work); -queue_work (priv->wq, &entry->work); +queue_work (entry->ct_priv->wq, &entry->work); } static struct mlx5_ct_counter *
@@ -1744,19 +1742,16 @@ mlx5_tc_ct_flush_ft_entry(void *ptr, void *arg)static void mlx5_tc_ct_del_ft_cb (struct mlx5_tc_ct_priv *ct_priv,struct mlx5_ct_ft *ft) { -struct mlx5e_priv *priv; -if (!refcount_dec_and_test (&ft->refcount))return ; +flush_workqueue (ct_priv->wq);nf_flow_table_offload_del_cb (ft->nf_ft, mlx5_tc_ct_block_flow_offload, ft);rhashtable_remove_fast (&ct_priv->zone_ht, &ft->node, zone_params);rhashtable_free_and_destroy (&ft->ct_entries_ht, mlx5_tc_ct_flush_ft_entry, ct_priv); - priv =netdev_priv (ct_priv->netdev); -flush_workqueue (priv->wq);mlx5_tc_ct_free_pre_ct_tables (ft);mapping_remove (ct_priv->zone_mapping, ft->zone_restore_id);kfree (ft);
@@ -2139,6 +2134,12 @@ mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains,if (rhashtable_init (&ct_priv->ct_tuples_nat_ht, &tuples_nat_ht_params)) goto err_ct_tuples_nat_ht; + ct_priv->wq =alloc_ordered_workqueue ("mlx5e_ct_priv_wq" ,0 ); +if (!ct_priv->wq) { + err = -ENOMEM ; + goto err_wq; + } + err =mlx5_tc_ct_fs_init (ct_priv);if (err) goto err_init_fs;
@@ -2146,6 +2147,8 @@ mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains,return ct_priv; err_init_fs: +destroy_workqueue (ct_priv->wq); +err_wq:rhashtable_destroy (&ct_priv->ct_tuples_nat_ht); err_ct_tuples_nat_ht:rhashtable_destroy (&ct_priv->ct_tuples_ht);
@@ -2175,6 +2178,7 @@ mlx5_tc_ct_clean(struct mlx5_tc_ct_priv *ct_priv)if (!ct_priv)return ; +destroy_workqueue (ct_priv->wq); chains = ct_priv->chains; ct_priv->fs_ops->destroy (ct_priv->fs);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 46/231 net/mlx5e: Fix capability check for updating vnic env counters
In reply to Greg Kroah-HartmanFrom: Gal Pressman <[email protected]>
[ Upstream commit 452133dd580811f184e76b1402983182ee425298 ]
The existing capability check for vnic env counters only checks for receive steering discards, although we need the counters update for the exposed internal queue oob counter as well. This could result in the latter counter not being updated correctly when the receive steering discards counter is not supported.
Fix that by checking whether any counter is supported instead of only the steering counter capability.
--- drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +1 -1
@@ -688,7 +688,7 @@ static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vnic_env)u32 in[MLX5_ST_SZ_DW (query_vnic_env_in)] = {};struct mlx5_core_dev *mdev = priv->mdev; -if (!MLX5_CAP_GEN (priv->mdev, nic_receive_steering_discard)) +if (!mlx5e_stats_grp_vnic_env_num_stats (priv))return ;MLX5_SET (query_vnic_env_in, in, opcode,MLX5_CMD_OP_QUERY_VNIC_ENV );
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 47/231 net/mlx5e: Ring the TX doorbell on DMA errors
In reply to Greg Kroah-HartmanFrom: Maxim Mikityanskiy <[email protected]>
[ Upstream commit 5b759bf2f9d73db05369aef2344502095c4e5e73 ]
TX doorbells may be postponed, because sometimes the driver knows that another packet follows (for example, when xmit_more is true, or when a MPWQE session is closed before transmitting a packet).
However, the DMA mapping may fail for the next packet, in which case a new WQE is not posted, the doorbell isn't updated either, and the transmission of the previous packet will be delayed indefinitely.
This commit fixes the described rare error flow by posting a NOP and ringing the doorbell on errors to flush all the previous packets. The MPWQE session is closed before that. DMA mapping in the MPWQE flow is moved to the beginning of mlx5e_sq_xmit_mpwqe, because empty sessions are not allowed. Stop room always has enough space for a NOP, because the actual TX WQE is not posted.
--- .../net/ethernet/mellanox/mlx5/core/en_tx.c | 39 ++++++++++++++----- 1 file changed, 30 insertions(+), 9 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +30 -9
@@ -318,6 +318,26 @@ static void mlx5e_tx_check_stop(struct mlx5e_txqsq *sq) } } +static void mlx5e_tx_flush (struct mlx5e_txqsq *sq) +{ +struct mlx5e_tx_wqe_info *wi; +struct mlx5e_tx_wqe *wqe; +u16 pi; + +/* Must not be called when a MPWQE session is active but empty. */ +mlx5e_tx_mpwqe_ensure_complete (sq); + + pi =mlx5_wq_cyc_ctr2ix (&sq->wq, sq->pc); + wi = &sq->db.wqe_info[pi]; + + *wi = (struct mlx5e_tx_wqe_info ) { + .num_wqebbs =1 , + }; + + wqe =mlx5e_post_nop (&sq->wq, sq->sqn, &sq->pc); +mlx5e_notify_hw (&sq->wq, sq->pc, sq->uar_map, &wqe->ctrl); +} + static inline void mlx5e_txwqe_complete(struct mlx5e_txqsq *sq, struct sk_buff *skb, const struct mlx5e_tx_attr *attr,
@@ -410,6 +430,7 @@ mlx5e_sq_xmit_wqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, err_drop: stats->dropped++;dev_kfree_skb_any (skb); +mlx5e_tx_flush (sq); }static bool mlx5e_tx_skb_supports_mpwqe (struct sk_buff *skb,struct mlx5e_tx_attr *attr)
@@ -511,6 +532,13 @@ mlx5e_sq_xmit_mpwqe(struct mlx5e_txqsq *sq, struct sk_buff *skb,struct mlx5_wqe_ctrl_seg *cseg;struct mlx5e_xmit_data txd; + txd.data = skb->data; + txd.len = skb->len; + + txd.dma_addr =dma_map_single (sq->pdev, txd.data, txd.len,DMA_TO_DEVICE ); +if (unlikely (dma_mapping_error (sq->pdev, txd.dma_addr))) + goto err_unmap; +if (!mlx5e_tx_mpwqe_session_is_active (sq)) {mlx5e_tx_mpwqe_session_start (sq, eseg); }else if (!mlx5e_tx_mpwqe_same_eseg (sq, eseg)) {
@@ -520,18 +548,9 @@ mlx5e_sq_xmit_mpwqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, sq->stats->xmit_more += xmit_more; - txd.data = skb->data; - txd.len = skb->len; - - txd.dma_addr =dma_map_single (sq->pdev, txd.data, txd.len,DMA_TO_DEVICE ); -if (unlikely (dma_mapping_error (sq->pdev, txd.dma_addr))) - goto err_unmap;mlx5e_dma_push (sq, txd.dma_addr, txd.len,MLX5E_DMA_MAP_SINGLE ); -mlx5e_skb_fifo_push (&sq->db.skb_fifo, skb); -mlx5e_tx_mpwqe_add_dseg (sq, &txd); -mlx5e_tx_skb_update_hwts_flags (skb);if (unlikely (mlx5e_tx_mpwqe_is_full (&sq->mpwqe, sq->max_sq_mpw_wqebbs))) {
@@ -553,6 +572,7 @@ mlx5e_sq_xmit_mpwqe(struct mlx5e_txqsq *sq, struct sk_buff *skb,mlx5e_dma_unmap_wqe_err (sq,1 ); sq->stats->dropped++;dev_kfree_skb_any (skb); +mlx5e_tx_flush (sq); }void mlx5e_tx_mpwqe_ensure_complete (struct mlx5e_txqsq *sq)
@@ -935,5 +955,6 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, err_drop: stats->dropped++;dev_kfree_skb_any (skb); + mlx5e_tx_flush(sq); }#endif
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 48/231 drm/amdgpu: keep fbdev buffers pinned during suspend
In reply to Greg Kroah-HartmanFrom: Alex Deucher <[email protected]>
[ Upstream commit f9a89117fbdc63c0d4ab63a8f3596a72c245bcfe ]
Was dropped when we converted to the generic helpers.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 25 +++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +21 -4
@@ -1547,6 +1547,21 @@ bool amdgpu_crtc_get_scanout_position(struct drm_crtc *crtc, stime, etime, mode); } +static bool +amdgpu_display_robj_is_fb (struct amdgpu_device *adev,struct amdgpu_bo *robj) +{ +struct drm_device *dev =adev_to_drm (adev); +struct drm_fb_helper *fb_helper = dev->fb_helper; + +if (!fb_helper || !fb_helper->buffer) +return false; + +if (gem_to_amdgpu_bo (fb_helper->buffer->gem) != robj) +return false; + +return true; +} +int amdgpu_display_suspend_helper (struct amdgpu_device *adev) {struct drm_device *dev =adev_to_drm (adev);
@@ -1582,10 +1597,12 @@ int amdgpu_display_suspend_helper(struct amdgpu_device *adev)continue ; } robj =gem_to_amdgpu_bo (fb->obj[0 ]); - r =amdgpu_bo_reserve (robj, true); -if (r ==0 ) { -amdgpu_bo_unpin (robj); -amdgpu_bo_unreserve (robj); +if (!amdgpu_display_robj_is_fb (adev, robj)) { + r =amdgpu_bo_reserve (robj, true); +if (r ==0 ) { +amdgpu_bo_unpin (robj); +amdgpu_bo_unreserve (robj); + } } }return 0 ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 49/231 drm/amdgpu/display: disable prefer_shadow for generic fb helpers
In reply to Greg Kroah-HartmanFrom: Alex Deucher <[email protected]>
[ Upstream commit 3a4b1cc28fbdc2325b3e3ed7d8024995a75f9216 ]
Seems to break hibernation. Disable for now until we can root cause it.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c | 3 ++- drivers/gpu/drm/amd/amdgpu/dce_v10_0.c | 3 ++- drivers/gpu/drm/amd/amdgpu/dce_v11_0.c | 3 ++- drivers/gpu/drm/amd/amdgpu/dce_v6_0.c | 3 ++- drivers/gpu/drm/amd/amdgpu/dce_v8_0.c | 3 ++- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 ++- 6 files changed, 12 insertions(+), 6 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c +2 -1
@@ -494,7 +494,8 @@ static int amdgpu_vkms_sw_init(void *handle)adev_to_drm (adev)->mode_config.max_height =YRES_MAX ;adev_to_drm (adev)->mode_config.preferred_depth =24 ; -adev_to_drm (adev)->mode_config.prefer_shadow =1 ; +/* disable prefer shadow for now due to hibernation issues */ +adev_to_drm (adev)->mode_config.prefer_shadow =0 ;adev_to_drm (adev)->mode_config.fb_base = adev->gmc.aper_base;
drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +2 -1
@@ -2796,7 +2796,8 @@ static int dce_v10_0_sw_init(void *handle)adev_to_drm (adev)->mode_config.max_height =16384 ;adev_to_drm (adev)->mode_config.preferred_depth =24 ; -adev_to_drm (adev)->mode_config.prefer_shadow =1 ; +/* disable prefer shadow for now due to hibernation issues */ +adev_to_drm (adev)->mode_config.prefer_shadow =0 ;adev_to_drm (adev)->mode_config.fb_modifiers_not_supported = true;
drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +2 -1
@@ -2914,7 +2914,8 @@ static int dce_v11_0_sw_init(void *handle)adev_to_drm (adev)->mode_config.max_height =16384 ;adev_to_drm (adev)->mode_config.preferred_depth =24 ; -adev_to_drm (adev)->mode_config.prefer_shadow =1 ; +/* disable prefer shadow for now due to hibernation issues */ +adev_to_drm (adev)->mode_config.prefer_shadow =0 ;adev_to_drm (adev)->mode_config.fb_modifiers_not_supported = true;
drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +2 -1
@@ -2673,7 +2673,8 @@ static int dce_v6_0_sw_init(void *handle)adev_to_drm (adev)->mode_config.max_width =16384 ;adev_to_drm (adev)->mode_config.max_height =16384 ;adev_to_drm (adev)->mode_config.preferred_depth =24 ; -adev_to_drm (adev)->mode_config.prefer_shadow =1 ; +/* disable prefer shadow for now due to hibernation issues */ +adev_to_drm (adev)->mode_config.prefer_shadow =0 ;adev_to_drm (adev)->mode_config.fb_modifiers_not_supported = true;adev_to_drm (adev)->mode_config.fb_base = adev->gmc.aper_base;
drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +2 -1
@@ -2693,7 +2693,8 @@ static int dce_v8_0_sw_init(void *handle)adev_to_drm (adev)->mode_config.max_height =16384 ;adev_to_drm (adev)->mode_config.preferred_depth =24 ; -adev_to_drm (adev)->mode_config.prefer_shadow =1 ; +/* disable prefer shadow for now due to hibernation issues */ +adev_to_drm (adev)->mode_config.prefer_shadow =0 ;adev_to_drm (adev)->mode_config.fb_modifiers_not_supported = true;
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +2 -1
@@ -3847,7 +3847,8 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev)adev_to_drm (adev)->mode_config.max_height =16384 ;adev_to_drm (adev)->mode_config.preferred_depth =24 ; -adev_to_drm (adev)->mode_config.prefer_shadow =1 ; +/* disable prefer shadow for now due to hibernation issues */ +adev_to_drm (adev)->mode_config.prefer_shadow =0 ;/* indicates support for immediate flip */ adev_to_drm (adev)->mode_config.async_page_flip = true;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 50/231 drm/i915: fix a possible refcount leak in intel_dp_add_mst_connector()
In reply to Greg Kroah-HartmanFrom: Hangyu Hua <[email protected]>
[ Upstream commit 85144df9ff4652816448369de76897c57cbb1b93 ]
If drm_connector_init fails, intel_connector_free will be called to take care of proper free. So it is necessary to drop the refcount of port before intel_connector_free.
--- drivers/gpu/drm/i915/display/intel_dp_mst.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/i915/display/intel_dp_mst.c +1 -0
@@ -841,6 +841,7 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo ret =drm_connector_init (dev, connector, &intel_dp_mst_connector_funcs, DRM_MODE_CONNECTOR_DisplayPort);if (ret) { +drm_dp_mst_put_port_malloc (port);intel_connector_free (intel_connector);return NULL ; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 51/231 drm/i915/guc: ADL-N should use the same GuC FW as ADL-S
In reply to Greg Kroah-HartmanFrom: Daniele Ceraolo Spurio <[email protected]>
[ Upstream commit 25c95bf494067f7bd1dfa8064ef964abe88cafc2 ]
The only difference between the ADL S and P GuC FWs is the HWConfig support. ADL-N does not support HWConfig, so we should use the same binary as ADL-S, otherwise the GuC might attempt to fetch a config table that does not exist. ADL-N is internally identified as an ADL-P, so we need to special-case it in the FW selection code.
--- drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +9 -0
@@ -160,6 +160,15 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw)u8 rev =INTEL_REVID (i915);int i; +/* +* The only difference between the ADL GuC FWs is the HWConfig support. +* ADL-N does not support HWConfig, so we should use the same binary as +* ADL-S, otherwise the GuC might attempt to fetch a config table that +* does not exist. +*/ +if (IS_ADLP_N (i915)) + p =INTEL_ALDERLAKE_S ; +GEM_BUG_ON (uc_fw->type >=ARRAY_SIZE (blobs_all)); fw_blobs = blobs_all[uc_fw->type].blobs; fw_count = blobs_all[uc_fw->type].count;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 52/231 ima: Fix a potential integer overflow in ima_appraise_measurement
In reply to Greg Kroah-HartmanFrom: Huaxin Lu <[email protected]>
[ Upstream commit d2ee2cfc4aa85ff6a2a3b198a3a524ec54e3d999 ]
When the ima-modsig is enabled, the rc passed to evm_verifyxattr() may be negative, which may cause the integer overflow problem.
--- security/integrity/ima/ima_appraise.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
security/integrity/ima/ima_appraise.c +2 -1
@@ -408,7 +408,8 @@ int ima_appraise_measurement(enum ima_hooks func, goto out; } - status =evm_verifyxattr (dentry,XATTR_NAME_IMA , xattr_value, rc, iint); + status =evm_verifyxattr (dentry,XATTR_NAME_IMA , xattr_value, + rc <0 ?0 : rc, iint);switch (status) {case INTEGRITY_PASS :case INTEGRITY_PASS_IMMUTABLE :
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 53/231 ASoC: sgtl5000: Fix noise on shutdown/remove
In reply to Greg Kroah-HartmanFrom: Francesco Dolcini <[email protected]>
[ Upstream commit 040e3360af3736348112d29425bf5d0be5b93115 ]
Put the SGTL5000 in a silent/safe state on shutdown/remove, this is required since the SGTL5000 produces a constant noise on its output after it is configured and its clock is removed. Without this change this is happening every time the module is unbound/removed or from reboot till the clock is enabled again.
The issue was experienced on both a Toradex Colibri/Apalis iMX6, but can be easily reproduced everywhere just playing something on the codec and after that removing/unbinding the driver.
--- sound/soc/codecs/sgtl5000.c | 9 +++++++++ sound/soc/codecs/sgtl5000.h | 1 + 2 files changed, 10 insertions(+)
sound/soc/codecs/sgtl5000.c +9 -0
@@ -1796,6 +1796,9 @@ static int sgtl5000_i2c_remove(struct i2c_client *client) {struct sgtl5000_priv *sgtl5000 =i2c_get_clientdata (client); +regmap_write (sgtl5000->regmap,SGTL5000_CHIP_DIG_POWER ,SGTL5000_DIG_POWER_DEFAULT ); +regmap_write (sgtl5000->regmap,SGTL5000_CHIP_ANA_POWER ,SGTL5000_ANA_POWER_DEFAULT ); +clk_disable_unprepare (sgtl5000->mclk);regulator_bulk_disable (sgtl5000->num_supplies, sgtl5000->supplies);regulator_bulk_free (sgtl5000->num_supplies, sgtl5000->supplies);
@@ -1803,6 +1806,11 @@ static int sgtl5000_i2c_remove(struct i2c_client *client)return 0 ; } +static void sgtl5000_i2c_shutdown (struct i2c_client *client) +{ +sgtl5000_i2c_remove (client); +} +static const struct i2c_device_id sgtl5000_id[] = { {"sgtl5000" ,0 }, {},
@@ -1823,6 +1831,7 @@ static struct i2c_driver sgtl5000_i2c_driver = { }, .probe = sgtl5000_i2c_probe, .remove = sgtl5000_i2c_remove, + .shutdown = sgtl5000_i2c_shutdown, .id_table = sgtl5000_id, };
sound/soc/codecs/sgtl5000.h +1 -0
@@ -80,6 +80,7 @@/* * SGTL5000_CHIP_DIG_POWER */ +#define SGTL5000_DIG_POWER_DEFAULT 0x0000#define SGTL5000_ADC_EN 0x0040#define SGTL5000_DAC_EN 0x0020#define SGTL5000_DAP_POWERUP 0x0010
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 54/231 ASoC: tas2764: Add post reset delays
In reply to Greg Kroah-HartmanFrom: Martin Povišer <[email protected]>
[ Upstream commit cd10bb89b0d57bca98eb75e0444854a1c129a14e ]
Make sure there is at least 1 ms delay from reset to first command as is specified in the datasheet. This is a fix similar to commit 307f31452078 ("ASoC: tas2770: Insert post reset delay").
--- sound/soc/codecs/tas2764.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
sound/soc/codecs/tas2764.c +8 -2
@@ -42,10 +42,12 @@ static void tas2764_reset(struct tas2764_priv *tas2764)gpiod_set_value_cansleep (tas2764->reset_gpio,0 );msleep (20 );gpiod_set_value_cansleep (tas2764->reset_gpio,1 ); +usleep_range (1000 ,2000 ); }snd_soc_component_write (tas2764->component,TAS2764_SW_RST ,TAS2764_RST ); +usleep_range (1000 ,2000 ); } static int tas2764_set_bias_level(struct snd_soc_component *component,
@@ -107,8 +109,10 @@ static int tas2764_codec_resume(struct snd_soc_component *component)struct tas2764_priv *tas2764 =snd_soc_component_get_drvdata (component);int ret; -if (tas2764->sdz_gpio) +if (tas2764->sdz_gpio) {gpiod_set_value_cansleep (tas2764->sdz_gpio,1 ); +usleep_range (1000 ,2000 ); + } ret = snd_soc_component_update_bits(component, TAS2764_PWR_CTRL, TAS2764_PWR_CTRL_MASK,
@@ -501,8 +505,10 @@ static int tas2764_codec_probe(struct snd_soc_component *component) tas2764->component = component; -if (tas2764->sdz_gpio) +if (tas2764->sdz_gpio) {gpiod_set_value_cansleep (tas2764->sdz_gpio,1 ); +usleep_range (1000 ,2000 ); + }tas2764_reset (tas2764);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 55/231 ASoC: tas2764: Fix and extend FSYNC polarity handling
In reply to Greg Kroah-HartmanFrom: Martin Povišer <[email protected]>
[ Upstream commit d1a10f1b48202e2d183cce144c218a211e98d906 ]
Fix setting of FSYNC polarity in case of LEFT_J and DSP_A/B formats. Do NOT set the SCFG field as was previously done, because that is not correct and is also in conflict with the "ASI1 Source" control which sets the same SCFG field!
Also add support for explicit polarity inversion.
--- sound/soc/codecs/tas2764.c | 30 +++++++++++++++++------------- sound/soc/codecs/tas2764.h | 6 ++---- 2 files changed, 19 insertions(+), 17 deletions(-)
sound/soc/codecs/tas2764.c +17 -13
@@ -135,7 +135,8 @@ static const char * const tas2764_ASI1_src[] = { };static SOC_ENUM_SINGLE_DECL ( - tas2764_ASI1_src_enum, TAS2764_TDM_CFG2, 4, tas2764_ASI1_src); + tas2764_ASI1_src_enum, TAS2764_TDM_CFG2, TAS2764_TDM_CFG2_SCFG_SHIFT, + tas2764_ASI1_src);static const struct snd_kcontrol_new tas2764_asi1_mux =SOC_DAPM_ENUM ("ASI1 Source" , tas2764_ASI1_src_enum);
@@ -333,20 +334,22 @@ static int tas2764_set_fmt(struct snd_soc_dai *dai, unsigned int fmt) {struct snd_soc_component *component = dai->component;struct tas2764_priv *tas2764 =snd_soc_component_get_drvdata (component); -u8 tdm_rx_start_slot =0 , asi_cfg_1 =0 ; -int iface; +u8 tdm_rx_start_slot =0 , asi_cfg_0 =0 , asi_cfg_1 =0 ;int ret;switch (fmt &SND_SOC_DAIFMT_INV_MASK ) { +case SND_SOC_DAIFMT_NB_IF : + asi_cfg_0 ^=TAS2764_TDM_CFG0_FRAME_START ; + fallthrough;case SND_SOC_DAIFMT_NB_NF : asi_cfg_1 =TAS2764_TDM_CFG1_RX_RISING ;break ; +case SND_SOC_DAIFMT_IB_IF : + asi_cfg_0 ^=TAS2764_TDM_CFG0_FRAME_START ; + fallthrough;case SND_SOC_DAIFMT_IB_NF : asi_cfg_1 =TAS2764_TDM_CFG1_RX_FALLING ;break ; -default : -dev_err (tas2764->dev,"ASI format Inverse is not found\n" ); -return -EINVAL ; } ret = snd_soc_component_update_bits(component, TAS2764_TDM_CFG1,
@@ -357,13 +360,13 @@ static int tas2764_set_fmt(struct snd_soc_dai *dai, unsigned int fmt)switch (fmt &SND_SOC_DAIFMT_FORMAT_MASK ) {case SND_SOC_DAIFMT_I2S : + asi_cfg_0 ^=TAS2764_TDM_CFG0_FRAME_START ; + fallthrough;case SND_SOC_DAIFMT_DSP_A : - iface =TAS2764_TDM_CFG2_SCFG_I2S ; tdm_rx_start_slot =1 ;break ;case SND_SOC_DAIFMT_DSP_B :case SND_SOC_DAIFMT_LEFT_J : - iface =TAS2764_TDM_CFG2_SCFG_LEFT_J ; tdm_rx_start_slot =0 ;break ;default :
@@ -372,14 +375,15 @@ static int tas2764_set_fmt(struct snd_soc_dai *dai, unsigned int fmt)return -EINVAL ; } - ret =snd_soc_component_update_bits (component,TAS2764_TDM_CFG1 , -TAS2764_TDM_CFG1_MASK , - (tdm_rx_start_slot <<TAS2764_TDM_CFG1_51_SHIFT )); + ret =snd_soc_component_update_bits (component,TAS2764_TDM_CFG0 , +TAS2764_TDM_CFG0_FRAME_START , + asi_cfg_0);if (ret <0 )return ret; - ret =snd_soc_component_update_bits (component,TAS2764_TDM_CFG2 , -TAS2764_TDM_CFG2_SCFG_MASK , iface); + ret =snd_soc_component_update_bits (component,TAS2764_TDM_CFG1 , +TAS2764_TDM_CFG1_MASK , + (tdm_rx_start_slot <<TAS2764_TDM_CFG1_51_SHIFT ));if (ret <0 )return ret;
sound/soc/codecs/tas2764.h +2 -4
@@ -47,6 +47,7 @@#define TAS2764_TDM_CFG0_MASK GENMASK(3, 1)#define TAS2764_TDM_CFG0_44_1_48KHZ BIT(3)#define TAS2764_TDM_CFG0_88_2_96KHZ (BIT(3) | BIT(1)) +#define TAS2764_TDM_CFG0_FRAME_START BIT(0)/* TDM Configuration Reg1 */ #define TAS2764_TDM_CFG1 TAS2764_REG(0X0, 0x09)
@@ -66,10 +67,7 @@#define TAS2764_TDM_CFG2_RXS_16BITS 0x0#define TAS2764_TDM_CFG2_RXS_24BITS BIT(0)#define TAS2764_TDM_CFG2_RXS_32BITS BIT(1) -#define TAS2764_TDM_CFG2_SCFG_MASK GENMASK(5, 4) -#define TAS2764_TDM_CFG2_SCFG_I2S 0x0 -#define TAS2764_TDM_CFG2_SCFG_LEFT_J BIT(4) -#define TAS2764_TDM_CFG2_SCFG_RIGHT_J BIT(5) +#define TAS2764_TDM_CFG2_SCFG_SHIFT 4/* TDM Configuration Reg3 */ #define TAS2764_TDM_CFG3 TAS2764_REG(0X0, 0x0c)
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 56/231 ASoC: tas2764: Correct playback volume range
In reply to Greg Kroah-HartmanFrom: Hector Martin <[email protected]>
[ Upstream commit 3e99e5697e1f7120b5abc755e8a560b22612d6ed ]
DVC value 0xc8 is -100dB and 0xc9 is mute; this needs to map to -100.5dB as far as the dB scale is concerned. Fix that and enable the mute flag, so alsamixer correctly shows the control as <0 dB .. -100 dB, mute>.
--- sound/soc/codecs/tas2764.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/codecs/tas2764.c +1 -1
@@ -536,7 +536,7 @@ static int tas2764_codec_probe(struct snd_soc_component *component) } static DECLARE_TLV_DB_SCALE(tas2764_digital_tlv, 1100, 50, 0); -static DECLARE_TLV_DB_SCALE(tas2764_playback_volume, -10000, 50, 0); +static DECLARE_TLV_DB_SCALE(tas2764_playback_volume, -10050, 50, 1);static const struct snd_kcontrol_new tas2764_snd_controls[] = { SOC_SINGLE_TLV("Speaker Volume", TAS2764_DVC, 0,
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 57/231 ASoC: tas2764: Fix amp gain register offset & default
In reply to Greg Kroah-HartmanFrom: Hector Martin <[email protected]>
[ Upstream commit 1c4f29ec878bbf1cc0a1eb54ae7da5ff98e19641 ]
The register default is 0x28 per the datasheet, and the amp gain field is supposed to be shifted left by one. With the wrong default, the ALSA controls lie about the power-up state. With the wrong shift, we get only half the gain we expect.
--- sound/soc/codecs/tas2764.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
sound/soc/codecs/tas2764.c +2 -2
@@ -541,7 +541,7 @@ static DECLARE_TLV_DB_SCALE(tas2764_playback_volume, -10050, 50, 1);static const struct snd_kcontrol_new tas2764_snd_controls[] = {SOC_SINGLE_TLV ("Speaker Volume" ,TAS2764_DVC ,0 ,TAS2764_DVC_MAX ,1 , tas2764_playback_volume), -SOC_SINGLE_TLV ("Amp Gain Volume" ,TAS2764_CHNL_0 ,0 ,0x14 ,0 , +SOC_SINGLE_TLV ("Amp Gain Volume" ,TAS2764_CHNL_0 ,1 ,0x14 ,0 , tas2764_digital_tlv), };
@@ -566,7 +566,7 @@ static const struct reg_default tas2764_reg_defaults[] = { { TAS2764_SW_RST, 0x00 }, { TAS2764_PWR_CTRL, 0x1a }, { TAS2764_DVC, 0x00 }, - { TAS2764_CHNL_0, 0x00 }, + { TAS2764_CHNL_0, 0x28 }, { TAS2764_TDM_CFG0, 0x09 }, { TAS2764_TDM_CFG1, 0x02 }, { TAS2764_TDM_CFG2, 0x0a },
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 58/231 ASoC: Intel: Skylake: Correct the ssp rate discovery in skl_get_ssp_clks()
In reply to Greg Kroah-HartmanFrom: Peter Ujfalusi <[email protected]>
[ Upstream commit 219af251bd1694bce1f627d238347d2eaf13de61 ]
The present flag is only set once when one rate has been found to be saved. This will effectively going to ignore any rate discovered at later time and based on the code, this is not the intention.
--- sound/soc/intel/skylake/skl-nhlt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
sound/soc/intel/skylake/skl-nhlt.c +2 -1
@@ -99,7 +99,6 @@ static void skl_get_ssp_clks(struct skl_dev *skl, struct skl_ssp_clk *ssp_clks,struct nhlt_fmt_cfg *fmt_cfg;struct wav_fmt_ext *wav_fmt;unsigned long rate; -bool present = false;int rate_index =0 ;u16 channels, bps;u8 clk_src;
@@ -113,6 +112,8 @@ static void skl_get_ssp_clks(struct skl_dev *skl, struct skl_ssp_clk *ssp_clks,return ;for (i =0 ; i < fmt->fmt_count; i++) { +bool present = false; + fmt_cfg = &fmt->fmt_config[i]; wav_fmt = &fmt_cfg->fmt_ext;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 59/231 ASoC: Intel: Skylake: Correct the handling of fmt_config flexible array
In reply to Greg Kroah-HartmanFrom: Peter Ujfalusi <[email protected]>
[ Upstream commit fc976f5629afb4160ee77798b14a693eac903ffd ]
The struct nhlt_format's fmt_config is a flexible array, it must not be used as normal array.
When moving to the next nhlt_fmt_cfg we need to take into account the data behind the ->config.caps (indicated by ->config.size).
The logic of the code also changed: it is no longer saves the _last_ fmt_cfg for all found rates.
--- sound/soc/intel/skylake/skl-nhlt.c | 37 ++++++++++++++++++++---------- 1 file changed, 25 insertions(+), 12 deletions(-)
sound/soc/intel/skylake/skl-nhlt.c +25 -12
@@ -111,11 +111,12 @@ static void skl_get_ssp_clks(struct skl_dev *skl, struct skl_ssp_clk *ssp_clks,if (fmt->fmt_count ==0 )return ; + fmt_cfg = (struct nhlt_fmt_cfg *)fmt->fmt_config;for (i =0 ; i < fmt->fmt_count; i++) { +struct nhlt_fmt_cfg *saved_fmt_cfg = fmt_cfg;bool present = false; - fmt_cfg = &fmt->fmt_config[i]; - wav_fmt = &fmt_cfg->fmt_ext; + wav_fmt = &saved_fmt_cfg->fmt_ext; channels = wav_fmt->fmt.channels; bps = wav_fmt->fmt.bits_per_sample;
@@ -133,12 +134,18 @@ static void skl_get_ssp_clks(struct skl_dev *skl, struct skl_ssp_clk *ssp_clks,* derive the rate. */ for (j = i; j < fmt->fmt_count; j++) { - fmt_cfg = &fmt->fmt_config[j]; - wav_fmt = &fmt_cfg->fmt_ext; +struct nhlt_fmt_cfg *tmp_fmt_cfg = fmt_cfg; + + wav_fmt = &tmp_fmt_cfg->fmt_ext;if ((fs == wav_fmt->fmt.samples_per_sec) && - (bps == wav_fmt->fmt.bits_per_sample)) + (bps == wav_fmt->fmt.bits_per_sample)) { channels =max_t (u16, channels, wav_fmt->fmt.channels); + saved_fmt_cfg = tmp_fmt_cfg; + } +/* Move to the next nhlt_fmt_cfg */ + tmp_fmt_cfg = (struct nhlt_fmt_cfg *)(tmp_fmt_cfg->config.caps + + tmp_fmt_cfg->config.size); } rate = channels * bps * fs;
@@ -154,8 +161,11 @@ static void skl_get_ssp_clks(struct skl_dev *skl, struct skl_ssp_clk *ssp_clks,/* Fill rate and parent for sclk/sclkfs */ if (!present) { +struct nhlt_fmt_cfg *first_fmt_cfg; + + first_fmt_cfg = (struct nhlt_fmt_cfg *)fmt->fmt_config; i2s_config_ext = (struct skl_i2s_config_blob_ext *) - fmt->fmt_config[0 ].config.caps; + first_fmt_cfg->config.caps;/* MCLK Divider Source Select */ if (is_legacy_blob (i2s_config_ext->hdr.sig)) {
@@ -169,6 +179,9 @@ static void skl_get_ssp_clks(struct skl_dev *skl, struct skl_ssp_clk *ssp_clks, parent =skl_get_parent_clk (clk_src); +/* Move to the next nhlt_fmt_cfg */ + fmt_cfg = (struct nhlt_fmt_cfg *)(fmt_cfg->config.caps + + fmt_cfg->config.size);/* * Do not copy the config data if there is no parent * clock available for this clock source select
@@ -177,9 +190,9 @@ static void skl_get_ssp_clks(struct skl_dev *skl, struct skl_ssp_clk *ssp_clks,continue ; sclk[id].rate_cfg[rate_index].rate = rate; - sclk[id].rate_cfg[rate_index].config = fmt_cfg; + sclk[id].rate_cfg[rate_index].config = saved_fmt_cfg; sclkfs[id].rate_cfg[rate_index].rate = rate; - sclkfs[id].rate_cfg[rate_index].config = fmt_cfg; + sclkfs[id].rate_cfg[rate_index].config = saved_fmt_cfg; sclk[id].parent_name = parent->name; sclkfs[id].parent_name = parent->name;
@@ -193,13 +206,13 @@ static void skl_get_mclk(struct skl_dev *skl, struct skl_ssp_clk *mclk, {struct skl_i2s_config_blob_ext *i2s_config_ext;struct skl_i2s_config_blob_legacy *i2s_config; -struct nhlt_specific_cfg *fmt_cfg; +struct nhlt_fmt_cfg *fmt_cfg;struct skl_clk_parent_src *parent;u32 clkdiv, div_ratio;u8 clk_src; - fmt_cfg = &fmt->fmt_config[0 ].config; - i2s_config_ext = (struct skl_i2s_config_blob_ext *)fmt_cfg->caps; + fmt_cfg = (struct nhlt_fmt_cfg *)fmt->fmt_config; + i2s_config_ext = (struct skl_i2s_config_blob_ext *)fmt_cfg->config.caps;/* MCLK Divider Source Select and divider */ if (is_legacy_blob (i2s_config_ext->hdr.sig)) {
@@ -228,7 +241,7 @@ static void skl_get_mclk(struct skl_dev *skl, struct skl_ssp_clk *mclk,return ; mclk[id].rate_cfg[0 ].rate = parent->rate/div_ratio; - mclk[id].rate_cfg[0 ].config = &fmt->fmt_config[0 ]; + mclk[id].rate_cfg[0 ].config = fmt_cfg; mclk[id].parent_name = parent->name; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 60/231 netfilter: ecache: move to separate structure
In reply to Greg Kroah-HartmanFrom: Florian Westphal <[email protected]>
[ Upstream commit 9027ce0b071a1bbd046682907fc2e23ca3592883 ]
This makes it easier for a followup patch to only expose ecache related parts of nf_conntrack_net structure.
--- include/net/netfilter/nf_conntrack.h | 8 ++++++-- net/netfilter/nf_conntrack_ecache.c | 19 ++++++++++--------- 2 files changed, 16 insertions(+), 11 deletions(-)
include/net/netfilter/nf_conntrack.h +6 -2
@@ -43,6 +43,11 @@ union nf_conntrack_expect_proto {/* insert expect proto private data here */ }; +struct nf_conntrack_net_ecache { +struct delayed_work dwork; +struct netns_ct *ct_net; +}; +struct nf_conntrack_net {/* only used when new connection is allocated: */ atomic_t count;
@@ -58,8 +63,7 @@ struct nf_conntrack_net {struct ctl_table_header *sysctl_header;#endif #ifdef CONFIG_NF_CONNTRACK_EVENTS -struct delayed_work ecache_dwork; -struct netns_ct *ct_net; +struct nf_conntrack_net_ecache ecache;#endif };
net/netfilter/nf_conntrack_ecache.c +10 -9
@@ -96,8 +96,8 @@ static enum retry_state ecache_work_evict_list(struct ct_pcpu *pcpu)static void ecache_work (struct work_struct *work) { - struct nf_conntrack_net *cnet = container_of(work, struct nf_conntrack_net, ecache_dwork.work); -struct netns_ct *ctnet = cnet->ct_net; + struct nf_conntrack_net *cnet = container_of(work, struct nf_conntrack_net, ecache.dwork.work); +struct netns_ct *ctnet = cnet->ecache.ct_net;int cpu, delay =-1 ;struct ct_pcpu *pcpu;
@@ -127,7 +127,7 @@ static void ecache_work(struct work_struct *work) ctnet->ecache_dwork_pending = delay >0 ;if (delay >=0 ) -schedule_delayed_work (&cnet->ecache_dwork, delay); +schedule_delayed_work (&cnet->ecache.dwork, delay); } static int __nf_conntrack_eventmask_report(struct nf_conntrack_ecache *e,
@@ -293,12 +293,12 @@ void nf_conntrack_ecache_work(struct net *net, enum nf_ct_ecache_state state)struct nf_conntrack_net *cnet =nf_ct_pernet (net);if (state ==NFCT_ECACHE_DESTROY_FAIL && - !delayed_work_pending (&cnet->ecache_dwork)) { -schedule_delayed_work (&cnet->ecache_dwork,HZ ); + !delayed_work_pending (&cnet->ecache.dwork)) { +schedule_delayed_work (&cnet->ecache.dwork,HZ ); net->ct.ecache_dwork_pending = true; }else if (state ==NFCT_ECACHE_DESTROY_SENT ) { net->ct.ecache_dwork_pending = false; -mod_delayed_work (system_wq, &cnet->ecache_dwork,0 ); +mod_delayed_work (system_wq, &cnet->ecache.dwork,0 ); } }
@@ -310,8 +310,9 @@ void nf_conntrack_ecache_pernet_init(struct net *net)struct nf_conntrack_net *cnet =nf_ct_pernet (net); net->ct.sysctl_events = nf_ct_events; - cnet->ct_net = &net->ct; -INIT_DELAYED_WORK (&cnet->ecache_dwork, ecache_work); + + cnet->ecache.ct_net = &net->ct; +INIT_DELAYED_WORK (&cnet->ecache.dwork, ecache_work);BUILD_BUG_ON (__IPCT_MAX >=16 );/* e->ctmask is u16 */ }
@@ -320,5 +321,5 @@ void nf_conntrack_ecache_pernet_fini(struct net *net) {struct nf_conntrack_net *cnet =nf_ct_pernet (net); -cancel_delayed_work_sync (&cnet->ecache_dwork); +cancel_delayed_work_sync (&cnet->ecache.dwork); }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 61/231 netfilter: conntrack: split inner loop of list dumping to own function
In reply to Greg Kroah-HartmanFrom: Florian Westphal <[email protected]>
[ Upstream commit 49001a2e83a80f6d9c4287c46ffa41a03667bbd1 ]
This allows code re-use in the followup patch. No functional changes intended.
--- net/netfilter/nf_conntrack_netlink.c | 68 ++++++++++++++++++---------- 1 file changed, 43 insertions(+), 25 deletions(-)
net/netfilter/nf_conntrack_netlink.c +43 -25
@@ -1708,6 +1708,47 @@ static int ctnetlink_done_list(struct netlink_callback *cb)return 0 ; } +static int ctnetlink_dump_one_entry (struct sk_buff *skb, +struct netlink_callback *cb, +struct nf_conn *ct, +bool dying) +{ +struct ctnetlink_list_dump_ctx *ctx = (void *)cb->ctx; +struct nfgenmsg *nfmsg =nlmsg_data (cb->nlh); +u8 l3proto = nfmsg->nfgen_family; +int res; + +if (l3proto &&nf_ct_l3num (ct) != l3proto) +return 0 ; + +if (ctx->last) { +if (ct != ctx->last) +return 0 ; + + ctx->last =NULL ; + } + +/* We can't dump extension info for the unconfirmed +* list because unconfirmed conntracks can have +* ct->ext reallocated (and thus freed). +* +* In the dying list case ct->ext can't be free'd +* until after we drop pcpu->lock. +*/ + res =ctnetlink_fill_info (skb,NETLINK_CB (cb->skb).portid, + cb->nlh->nlmsg_seq, +NFNL_MSG_TYPE (cb->nlh->nlmsg_type), + ct, dying,0 ); +if (res <0 ) { +if (!refcount_inc_not_zero (&ct->ct_general.use)) +return 0 ; + + ctx->last = ct; + } + +return res; +} +static int ctnetlink_dump_list (struct sk_buff *skb,struct netlink_callback *cb,bool dying) {
@@ -1715,12 +1756,9 @@ ctnetlink_dump_list(struct sk_buff *skb, struct netlink_callback *cb, bool dyingstruct nf_conn *ct, *last;struct nf_conntrack_tuple_hash *h;struct hlist_nulls_node *n; -struct nfgenmsg *nfmsg =nlmsg_data (cb->nlh); -u_int8_t l3proto = nfmsg->nfgen_family; -int res; -int cpu;struct hlist_nulls_head *list;struct net *net =sock_net (skb->sk); +int res, cpu;if (ctx->done)return 0 ;
@@ -1739,30 +1777,10 @@ ctnetlink_dump_list(struct sk_buff *skb, struct netlink_callback *cb, bool dying restart:hlist_nulls_for_each_entry (h, n, list, hnnode) { ct =nf_ct_tuplehash_to_ctrack (h); -if (l3proto &&nf_ct_l3num (ct) != l3proto) -continue ; -if (ctx->last) { -if (ct != last) -continue ; - ctx->last =NULL ; - } -/* We can't dump extension info for the unconfirmed -* list because unconfirmed conntracks can have -* ct->ext reallocated (and thus freed). -* -* In the dying list case ct->ext can't be free'd -* until after we drop pcpu->lock. -*/ - res =ctnetlink_fill_info (skb,NETLINK_CB (cb->skb).portid, - cb->nlh->nlmsg_seq, -NFNL_MSG_TYPE (cb->nlh->nlmsg_type), - ct, dying,0 ); + res =ctnetlink_dump_one_entry (skb, cb, ct, dying);if (res <0 ) { -if (!refcount_inc_not_zero (&ct->ct_general.use)) -continue ; ctx->cpu = cpu; - ctx->last = ct;spin_unlock_bh (&pcpu->lock); goto out; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 62/231 netfilter: ecache: use dedicated list for event redelivery
In reply to Greg Kroah-HartmanFrom: Florian Westphal <[email protected]>
[ Upstream commit 2ed3bf188b33630cf9d93b996ebf001847a00b5a ]
This disentangles event redelivery and the percpu dying list.
Because entries are now stored on a dedicated list, all entries are in NFCT_ECACHE_DESTROY_FAIL state and all entries still have confirmed bit set -- the reference count is at least 1.
The 'struct net' back-pointer can be removed as well.
The pcpu dying list will be removed eventually, it has no functionality.
--- include/net/netfilter/nf_conntrack.h | 3 +- include/net/netfilter/nf_conntrack_ecache.h | 2 - net/netfilter/nf_conntrack_core.c | 33 +++++- net/netfilter/nf_conntrack_ecache.c | 117 +++++++++----------- 4 files changed, 82 insertions(+), 73 deletions(-)
include/net/netfilter/nf_conntrack.h +2 -1
@@ -45,7 +45,8 @@ union nf_conntrack_expect_proto {struct nf_conntrack_net_ecache {struct delayed_work dwork; -struct netns_ct *ct_net; +spinlock_t dying_lock; +struct hlist_nulls_head dying_list; };struct nf_conntrack_net {
include/net/netfilter/nf_conntrack_ecache.h +0 -2
@@ -14,7 +14,6 @@#include <net/netfilter/nf_conntrack_extend.h> enum nf_ct_ecache_state { -NFCT_ECACHE_UNKNOWN ,/* destroy event not sent */ NFCT_ECACHE_DESTROY_FAIL ,/* tried but failed to send destroy event */ NFCT_ECACHE_DESTROY_SENT ,/* sent destroy event after failure */ };
@@ -23,7 +22,6 @@ struct nf_conntrack_ecache {unsigned long cache;/* bitops want long */ u16 ctmask;/* bitmask of ct events to be delivered */ u16 expmask;/* bitmask of expect events to be delivered */ - enum nf_ct_ecache_state state:8;/* ecache state */u32 missed;/* missed events */ u32 portid;/* netlink portid of destroyer */ };
net/netfilter/nf_conntrack_core.c +28 -5
@@ -660,15 +660,12 @@ void nf_ct_destroy(struct nf_conntrack *nfct) }EXPORT_SYMBOL (nf_ct_destroy); -static void nf_ct_delete_from_lists(struct nf_conn *ct) +static void __nf_ct_delete_from_lists(struct nf_conn *ct) { struct net *net = nf_ct_net(ct); unsigned int hash, reply_hash; unsigned int sequence; - nf_ct_helper_destroy(ct); - - local_bh_disable(); do { sequence = read_seqcount_begin(&nf_conntrack_generation); hash = hash_conntrack(net,
@@ -681,12 +678,33 @@ static void nf_ct_delete_from_lists(struct nf_conn *ct)clean_from_lists (ct);nf_conntrack_double_unlock (hash, reply_hash); +} +static void nf_ct_delete_from_lists (struct nf_conn *ct) +{ +nf_ct_helper_destroy (ct); +local_bh_disable (); + +__nf_ct_delete_from_lists (ct);nf_ct_add_to_dying_list (ct);local_bh_enable (); } +static void nf_ct_add_to_ecache_list (struct nf_conn *ct) +{ +#ifdef CONFIG_NF_CONNTRACK_EVENTS +struct nf_conntrack_net *cnet =nf_ct_pernet (nf_ct_net (ct)); + +spin_lock (&cnet->ecache.dying_lock); +hlist_nulls_add_head_rcu (&ct->tuplehash[IP_CT_DIR_ORIGINAL ].hnnode, + &cnet->ecache.dying_list); +spin_unlock (&cnet->ecache.dying_lock); +#else +nf_ct_add_to_dying_list (ct); +#endif +} +bool nf_ct_delete (struct nf_conn *ct,u32 portid,int report) {struct nf_conn_tstamp *tstamp;
@@ -709,7 +727,12 @@ bool nf_ct_delete(struct nf_conn *ct, u32 portid, int report)/* destroy event was not delivered. nf_ct_put will * be done by event cache worker on redelivery. */ -nf_ct_delete_from_lists (ct); +nf_ct_helper_destroy (ct); +local_bh_disable (); +__nf_ct_delete_from_lists (ct); +nf_ct_add_to_ecache_list (ct); +local_bh_enable (); +nf_conntrack_ecache_work (nf_ct_net (ct),NFCT_ECACHE_DESTROY_FAIL );return false; }
net/netfilter/nf_conntrack_ecache.c +52 -65
183 lines not on this page · raw message
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 63/231 netfilter: conntrack: include ecache dying list in dumps
In reply to Greg Kroah-HartmanFrom: Florian Westphal <[email protected]>
[ Upstream commit 0d3cc504ba9cdcff76346306c37eb1ea01e60a86 ]
The new pernet dying list includes conntrack entries that await delivery of the 'destroy' event via ctnetlink.
The old percpu dying list will be removed soon.
--- include/net/netfilter/nf_conntrack_ecache.h | 2 + net/netfilter/nf_conntrack_ecache.c | 10 +++++ net/netfilter/nf_conntrack_netlink.c | 43 +++++++++++++++++++++ 3 files changed, 55 insertions(+)
include/net/netfilter/nf_conntrack_ecache.h +2 -0
@@ -164,6 +164,8 @@ void nf_conntrack_ecache_work(struct net *net, enum nf_ct_ecache_state state);void nf_conntrack_ecache_pernet_init (struct net *net);void nf_conntrack_ecache_pernet_fini (struct net *net); +struct nf_conntrack_net_ecache *nf_conn_pernet_ecache (const struct net *net); +static inline bool nf_conntrack_ecache_dwork_pending (const struct net *net) {return net->ct.ecache_dwork_pending;
net/netfilter/nf_conntrack_ecache.c +10 -0
@@ -38,6 +38,16 @@ enum retry_state { STATE_DONE, }; +struct nf_conntrack_net_ecache *nf_conn_pernet_ecache (const struct net *net) +{ +struct nf_conntrack_net *cnet =nf_ct_pernet (net); + +return &cnet->ecache; +} +#if IS_MODULE (CONFIG_NF_CT_NETLINK ) +EXPORT_SYMBOL_GPL (nf_conn_pernet_ecache); +#endif +static enum retry_state ecache_work_evict_list (struct nf_conntrack_net *cnet) {unsigned long stop = jiffies +ECACHE_MAX_JIFFIES ;
net/netfilter/nf_conntrack_netlink.c +43 -0
55 lines not on this page · raw message
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 64/231 netfilter: conntrack: remove the percpu dying list
In reply to Greg Kroah-HartmanFrom: Florian Westphal <[email protected]>
[ Upstream commit 1397af5bfd7d32b0cf2adb70a78c9a9e8f11d912 ]
Its no longer needed. Entries that need event redelivery are placed on the new pernet dying list.
The advantage is that there is no need to take additional spinlock on conntrack removal unless event redelivery failed or the conntrack entry was never added to the table in the first place (confirmed bit not set).
The IPS_CONFIRMED bit now needs to be set as soon as the entry has been unlinked from the unconfirmed list, else the destroy function may attempt to unlink it a second time.
--- include/net/netns/conntrack.h | 1 - net/netfilter/nf_conntrack_core.c | 35 +++++----------------------- net/netfilter/nf_conntrack_ecache.c | 1 - net/netfilter/nf_conntrack_netlink.c | 23 ++++++------------ 4 files changed, 13 insertions(+), 47 deletions(-)
include/net/netns/conntrack.h +0 -1
@@ -96,7 +96,6 @@ struct nf_ip_net {struct ct_pcpu {spinlock_t lock;struct hlist_nulls_head unconfirmed; -struct hlist_nulls_head dying; };struct netns_ct {
net/netfilter/nf_conntrack_core.c +6 -29
@@ -525,21 +525,6 @@ clean_from_lists(struct nf_conn *ct)nf_ct_remove_expectations (ct); } -/* must be called with local_bh_disable */ -static void nf_ct_add_to_dying_list (struct nf_conn *ct) -{ -struct ct_pcpu *pcpu; - -/* add this conntrack to the (per cpu) dying list */ - ct->cpu =smp_processor_id (); - pcpu =per_cpu_ptr (nf_ct_net (ct)->ct.pcpu_lists, ct->cpu); - -spin_lock (&pcpu->lock); -hlist_nulls_add_head (&ct->tuplehash[IP_CT_DIR_ORIGINAL ].hnnode, - &pcpu->dying); -spin_unlock (&pcpu->lock); -} -/* must be called with local_bh_disable */ static void nf_ct_add_to_unconfirmed_list (struct nf_conn *ct) {
@@ -556,11 +541,11 @@ static void nf_ct_add_to_unconfirmed_list(struct nf_conn *ct) }/* must be called with local_bh_disable */ -static void nf_ct_del_from_dying_or_unconfirmed_list (struct nf_conn *ct) +static void nf_ct_del_from_unconfirmed_list (struct nf_conn *ct) {struct ct_pcpu *pcpu; -/* We overload first tuple to link into unconfirmed or dying list.*/ +/* We overload first tuple to link into unconfirmed list.*/ pcpu =per_cpu_ptr (nf_ct_net (ct)->ct.pcpu_lists, ct->cpu);spin_lock (&pcpu->lock);
@@ -648,7 +633,8 @@ void nf_ct_destroy(struct nf_conntrack *nfct)*/ nf_ct_remove_expectations (ct); -nf_ct_del_from_dying_or_unconfirmed_list (ct); +if (unlikely (!nf_ct_is_confirmed (ct))) +nf_ct_del_from_unconfirmed_list (ct);local_bh_enable ();
@@ -686,7 +672,6 @@ static void nf_ct_delete_from_lists(struct nf_conn *ct)local_bh_disable ();__nf_ct_delete_from_lists (ct); -nf_ct_add_to_dying_list (ct);local_bh_enable (); }
@@ -700,8 +685,6 @@ static void nf_ct_add_to_ecache_list(struct nf_conn *ct)hlist_nulls_add_head_rcu (&ct->tuplehash[IP_CT_DIR_ORIGINAL ].hnnode, &cnet->ecache.dying_list);spin_unlock (&cnet->ecache.dying_lock); -#else -nf_ct_add_to_dying_list (ct);#endif }
@@ -995,7 +978,6 @@ static void __nf_conntrack_insert_prepare(struct nf_conn *ct)struct nf_conn_tstamp *tstamp;refcount_inc (&ct->ct_general.use); - ct->status |=IPS_CONFIRMED ;/* set conntrack timestamp, if enabled. */ tstamp =nf_conn_tstamp_find (ct);
@@ -1024,7 +1006,6 @@ static int __nf_ct_resolve_clash(struct sk_buff *skb,nf_conntrack_get (&ct->ct_general);nf_ct_acct_merge (ct, ctinfo, loser_ct); -nf_ct_add_to_dying_list (loser_ct);nf_ct_put (loser_ct);nf_ct_set (skb, ct, ctinfo);
@@ -1157,7 +1138,6 @@ nf_ct_resolve_clash(struct sk_buff *skb, struct nf_conntrack_tuple_hash *h,return ret; drop: -nf_ct_add_to_dying_list (loser_ct);NF_CT_STAT_INC (net, drop);NF_CT_STAT_INC (net, insert_failed);return NF_DROP ;
@@ -1224,10 +1204,10 @@ __nf_conntrack_confirm(struct sk_buff *skb)* user context, else we insert an already 'dead' hash, blocking * further use of that particular connection -JM. */ -nf_ct_del_from_dying_or_unconfirmed_list (ct); +nf_ct_del_from_unconfirmed_list (ct); + ct->status |=IPS_CONFIRMED ;if (unlikely (nf_ct_is_dying (ct))) { -nf_ct_add_to_dying_list (ct);NF_CT_STAT_INC (net, insert_failed); goto dying; }
@@ -1251,7 +1231,6 @@ __nf_conntrack_confirm(struct sk_buff *skb) goto out;if (chainlen++ > max_chainlen) { chaintoolong: -nf_ct_add_to_dying_list (ct);NF_CT_STAT_INC (net, chaintoolong);NF_CT_STAT_INC (net, insert_failed); ret =NF_DROP ;
@@ -2800,7 +2779,6 @@ void nf_conntrack_init_end(void)* We need to use special "null" values, not used in hash table */ #define UNCONFIRMED_NULLS_VAL ((1<<30)+0) -#define DYING_NULLS_VAL ((1<<30)+1)int nf_conntrack_init_net (struct net *net) {
@@ -2821,7 +2799,6 @@ int nf_conntrack_init_net(struct net *net)spin_lock_init (&pcpu->lock);INIT_HLIST_NULLS_HEAD (&pcpu->unconfirmed,UNCONFIRMED_NULLS_VAL ); -INIT_HLIST_NULLS_HEAD (&pcpu->dying,DYING_NULLS_VAL ); } net->ct.stat = alloc_percpu(struct ip_conntrack_stat);
net/netfilter/nf_conntrack_ecache.c +0 -1
@@ -94,7 +94,6 @@ static enum retry_state ecache_work_evict_list(struct nf_conntrack_net *cnet)hlist_nulls_for_each_entry_safe (h, n, &evicted_list, hnnode) {struct nf_conn *ct =nf_ct_tuplehash_to_ctrack (h); -hlist_nulls_add_fake (&ct->tuplehash[IP_CT_DIR_ORIGINAL ].hnnode);hlist_nulls_del_rcu (&ct->tuplehash[IP_CT_DIR_REPLY ].hnnode);nf_ct_put (ct);
net/netfilter/nf_conntrack_netlink.c +7 -16
@@ -62,7 +62,6 @@ struct ctnetlink_list_dump_ctx {struct nf_conn *last;unsigned int cpu;bool done; -bool retrans_done; }; static int ctnetlink_dump_tuples_proto(struct sk_buff *skb,
@@ -1751,13 +1750,12 @@ static int ctnetlink_dump_one_entry(struct sk_buff *skb, }static int -ctnetlink_dump_list (struct sk_buff *skb,struct netlink_callback *cb,bool dying) +ctnetlink_dump_unconfirmed (struct sk_buff *skb,struct netlink_callback *cb) {struct ctnetlink_list_dump_ctx *ctx = (void *)cb->ctx;struct nf_conn *ct, *last;struct nf_conntrack_tuple_hash *h;struct hlist_nulls_node *n; -struct hlist_nulls_head *list;struct net *net =sock_net (skb->sk);int res, cpu;
@@ -1774,12 +1772,11 @@ ctnetlink_dump_list(struct sk_buff *skb, struct netlink_callback *cb, bool dying pcpu =per_cpu_ptr (net->ct.pcpu_lists, cpu);spin_lock_bh (&pcpu->lock); - list = dying ? &pcpu->dying : &pcpu->unconfirmed; restart: -hlist_nulls_for_each_entry (h, n, list, hnnode) { +hlist_nulls_for_each_entry (h, n, &pcpu->unconfirmed, hnnode) { ct =nf_ct_tuplehash_to_ctrack (h); - res =ctnetlink_dump_one_entry (skb, cb, ct, dying); + res =ctnetlink_dump_one_entry (skb, cb, ct, false);if (res <0 ) { ctx->cpu = cpu;spin_unlock_bh (&pcpu->lock);
@@ -1812,8 +1809,8 @@ ctnetlink_dump_dying(struct sk_buff *skb, struct netlink_callback *cb)struct hlist_nulls_node *n;#endif -if (ctx->retrans_done) -return ctnetlink_dump_list (skb, cb, true); +if (ctx->done) +return 0 ; ctx->last =NULL ;
@@ -1842,10 +1839,10 @@ ctnetlink_dump_dying(struct sk_buff *skb, struct netlink_callback *cb)spin_unlock_bh (&ecache_net->dying_lock);#endif + ctx->done = true;nf_ct_put (last); - ctx->retrans_done = true; -return ctnetlink_dump_list (skb, cb, true); +return skb->len; } static int ctnetlink_get_ct_dying(struct sk_buff *skb,
@@ -1863,12 +1860,6 @@ static int ctnetlink_get_ct_dying(struct sk_buff *skb,return -EOPNOTSUPP ; } -static int -ctnetlink_dump_unconfirmed (struct sk_buff *skb,struct netlink_callback *cb) -{ -return ctnetlink_dump_list (skb, cb, false); -} -static int ctnetlink_get_ct_unconfirmed (struct sk_buff *skb,const struct nfnl_info *info,const struct nlattr *const cda[])
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 65/231 netfilter: conntrack: fix crash due to confirmed bit load reordering
In reply to Greg Kroah-HartmanFrom: Florian Westphal <[email protected]>
[ Upstream commit 0ed8f619b412b52360ccdfaf997223ccd9319569 ]
Kajetan Puchalski reports crash on ARM, with backtrace of:
__nf_ct_delete_from_listsnf_ct_deleteearly_drop__nf_conntrack_alloc
Unlike atomic_inc_not_zero, refcount_inc_not_zero is not a full barrier. conntrack uses SLAB_TYPESAFE_BY_RCU, i.e. it is possible that a 'newly' allocated object is still in use on another CPU:
CPU1 CPU2
encounter 'ct' during hlist walk
delete_from_lists
refcount drops to 0
kmem_cache_free(ct);
__nf_conntrack_alloc() // returns same object
refcount_inc_not_zero(ct); /* might fail */ /* If set, ct is public/in the hash table */ test_bit(IPS_CONFIRMED_BIT, &ct->status);
In case CPU1 already set refcount back to 1, refcount_inc_not_zero() will succeed.
The expected possibilities for a CPU that obtained the object 'ct' (but no reference so far) are:
1. refcount_inc_not_zero() fails. CPU2 ignores the object and moves to the next entry in the list. This happens for objects that are about to be free'd, that have been free'd, or that have been reallocated by __nf_conntrack_alloc(), but where the refcount has not been increased back to 1 yet.
2. refcount_inc_not_zero() succeeds. CPU2 checks the CONFIRMED bit in ct->status. If set, the object is public/in the table.
If not, the object must be skipped; CPU2 calls nf_ct_put() to un-do the refcount increment and moves to the next object.
Parallel deletion from the hlists is prevented by a 'test_and_set_bit(IPS_DYING_BIT, &ct->status);' check, i.e. only one cpu will do the unlink, the other one will only drop its reference count.
Because refcount_inc_not_zero is not a full barrier, CPU2 may try to delete an object that is not on any list:
1. refcount_inc_not_zero() successful (refcount inited to 1 on other CPU)
2. CONFIRMED test also successful (load was reordered or zeroing of ct->status not yet visible)
3. delete_from_lists unlinks entry not on the hlist, because IPS_DYING_BIT is 0 (already cleared).
2) is already wrong: CPU2 will handle a partially initited object that is supposed to be private to CPU1.
Add needed barriers when refcount_inc_not_zero() is successful.
It also inserts a smp_wmb() before the refcount is set to 1 during allocation.
Because other CPU might still see the object, refcount_set(1) "resurrects" it, so we need to make sure that other CPUs will also observe the right content. In particular, the CONFIRMED bit test must only pass once the object is fully initialised and either in the hash or about to be inserted (with locks held to delay possible unlink from early_drop or gc worker).
I did not change flow_offload_alloc(), as far as I can see it should call refcount_inc(), not refcount_inc_not_zero(): the ct object is attached to the skb so its refcount should be >= 1 in all cases.
v2: prefer smp_acquire__after_ctrl_dep to smp_rmb (Will Deacon).
v3: keep smp_acquire__after_ctrl_dep close to refcount_inc_not_zero call
add comment in nf_conntrack_netlink, no control dependency there due to locks.
--- net/netfilter/nf_conntrack_core.c | 22 ++++++++++++++++++++++ net/netfilter/nf_conntrack_netlink.c | 1 + net/netfilter/nf_conntrack_standalone.c | 3 +++ 3 files changed, 26 insertions(+)
net/netfilter/nf_conntrack_core.c +22 -0
@@ -764,6 +764,9 @@ static void nf_ct_gc_expired(struct nf_conn *ct)if (!refcount_inc_not_zero (&ct->ct_general.use))return ; +/* load ->status after refcount increase */ +smp_acquire__after_ctrl_dep (); +if (nf_ct_should_gc (ct))nf_ct_kill (ct);
@@ -830,6 +833,9 @@ __nf_conntrack_find_get(struct net *net, const struct nf_conntrack_zone *zone,*/ ct =nf_ct_tuplehash_to_ctrack (h);if (likely (refcount_inc_not_zero (&ct->ct_general.use))) { +/* re-check key after refcount */ +smp_acquire__after_ctrl_dep (); +if (likely (nf_ct_key_equal (h, tuple, zone, net))) goto found;
@@ -1369,6 +1375,9 @@ static unsigned int early_drop_list(struct net *net,if (!refcount_inc_not_zero (&tmp->ct_general.use))continue ; +/* load ->ct_net and ->status after refcount increase */ +smp_acquire__after_ctrl_dep (); +/* kill only if still in same netns -- might have moved due to * SLAB_TYPESAFE_BY_RCU rules. *
@@ -1518,6 +1527,9 @@ static void gc_worker(struct work_struct *work)if (!refcount_inc_not_zero (&tmp->ct_general.use))continue ; +/* load ->status after refcount increase */ +smp_acquire__after_ctrl_dep (); +if (gc_worker_skip_ct (tmp)) {nf_ct_put (tmp);continue ;
@@ -1749,6 +1761,16 @@ init_conntrack(struct net *net, struct nf_conn *tmpl,if (!exp)__nf_ct_try_assign_helper (ct, tmpl,GFP_ATOMIC ); +/* Other CPU might have obtained a pointer to this object before it was +* released. Because refcount is 0, refcount_inc_not_zero() will fail. +* +* After refcount_set(1) it will succeed; ensure that zeroing of +* ct->status and the correct ct->net pointer are visible; else other +* core might observe CONFIRMED bit which means the entry is valid and +* in the hash table, but its not (anymore). +*/ +smp_wmb (); +/* Now it is inserted into the unconfirmed list, set refcount to 1. */ refcount_set (&ct->ct_general.use,1 );nf_ct_add_to_unconfirmed_list (ct);
net/netfilter/nf_conntrack_netlink.c +1 -0
@@ -1203,6 +1203,7 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) hnnode) { ct =nf_ct_tuplehash_to_ctrack (h);if (nf_ct_is_expired (ct)) { +/* need to defer nf_ct_kill() until lock is released */ if (i <ARRAY_SIZE (nf_ct_evict) &&refcount_inc_not_zero (&ct->ct_general.use)) nf_ct_evict[i++] = ct;
net/netfilter/nf_conntrack_standalone.c +3 -0
@@ -306,6 +306,9 @@ static int ct_seq_show(struct seq_file *s, void *v)if (unlikely (!refcount_inc_not_zero (&ct->ct_general.use)))return 0 ; +/* load ->status after refcount increase */ +smp_acquire__after_ctrl_dep (); +if (nf_ct_should_gc (ct)) {nf_ct_kill (ct); goto release;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 66/231 net: stmmac: dwc-qos: Disable split header for Tegra194
In reply to Greg Kroah-HartmanFrom: Jon Hunter <[email protected]>
[ Upstream commit 029c1c2059e9c4b38f97a06204cdecd10cfbeb8a ]
There is a long-standing issue with the Synopsys DWC Ethernet driver for Tegra194 where random system crashes have been observed [0]. The problem occurs when the split header feature is enabled in the stmmac driver. In the bad case, a larger than expected buffer length is received and causes the calculation of the total buffer length to overflow. This results in a very large buffer length that causes the kernel to crash. Why this larger buffer length is received is not clear, however, the feedback from the NVIDIA design team is that the split header feature is not supported for Tegra194. Therefore, disable split header support for Tegra194 to prevent these random crashes from occurring.
[0] https://lore.kernel.org/linux-tegra/[email protected]/
--- drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c +1 -0
@@ -361,6 +361,7 @@ static int tegra_eqos_probe(struct platform_device *pdev, data->fix_mac_speed = tegra_eqos_fix_speed; data->init = tegra_eqos_init; data->bsp_priv = eqos; + data->sph_disable =1 ; err =tegra_eqos_init (pdev, eqos);if (err <0 )
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 67/231 net: ethernet: ti: am65-cpsw: Fix devlink port register sequence
In reply to Greg Kroah-HartmanFrom: Siddharth Vadapalli <[email protected]>
[ Upstream commit 0680e20af5fbf41df8a11b11bd9a7c25b2ca0746 ]
Renaming interfaces using udevd depends on the interface being registered before its netdev is registered. Otherwise, udevd reads an empty phys_port_name value, resulting in the interface not being renamed.
Fix this by registering the interface before registering its netdev by invoking am65_cpsw_nuss_register_devlink() before invoking register_netdev() for the interface.
Move the function call to devlink_port_type_eth_set(), invoking it after register_netdev() is invoked, to ensure that netlink notification for the port state change is generated after the netdev is completely initialized.
--- drivers/net/ethernet/ti/am65-cpsw-nuss.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-)
drivers/net/ethernet/ti/am65-cpsw-nuss.c +10 -7
@@ -2475,7 +2475,6 @@ static int am65_cpsw_nuss_register_devlink(struct am65_cpsw_common *common) port->port_id, ret); goto dl_port_unreg; } -devlink_port_type_eth_set (dl_port, port->ndev); }devlink_register (common->devlink);return ret;
@@ -2519,6 +2518,7 @@ static void am65_cpsw_unregister_devlink(struct am65_cpsw_common *common)static int am65_cpsw_nuss_register_ndevs (struct am65_cpsw_common *common) {struct device *dev = common->dev; +struct devlink_port *dl_port;struct am65_cpsw_port *port;int ret =0 , i;
@@ -2535,6 +2535,10 @@ static int am65_cpsw_nuss_register_ndevs(struct am65_cpsw_common *common)return ret; } + ret =am65_cpsw_nuss_register_devlink (common); +if (ret) +return ret; +for (i =0 ; i < common->port_num; i++) { port = &common->ports[i];
@@ -2547,25 +2551,24 @@ static int am65_cpsw_nuss_register_ndevs(struct am65_cpsw_common *common) i, ret); goto err_cleanup_ndev; } + + dl_port = &port->devlink_port; +devlink_port_type_eth_set (dl_port, port->ndev); } ret =am65_cpsw_register_notifiers (common);if (ret) goto err_cleanup_ndev; - ret =am65_cpsw_nuss_register_devlink (common); -if (ret) - goto clean_unregister_notifiers; -/* can't auto unregister ndev using devm_add_action() due to * devres release sequence in DD core for DMA */ return 0 ; -clean_unregister_notifiers: -am65_cpsw_unregister_notifiers (common); + err_cleanup_ndev:am65_cpsw_nuss_cleanup_ndev (common); +am65_cpsw_unregister_devlink (common);return ret; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 68/231 net: ocelot: fix wrong time_after usage
In reply to Greg Kroah-HartmanFrom: Pavel Skripkin <[email protected]>
[ Upstream commit f46fd3d7c3bd5d7bd5bb664135cf32ca9e97190b ]
Accidentally noticed, that this driver is the only user of while (time_after(jiffies...)).
It looks like typo, because likely this while loop will finish after 1st iteration, because time_after() returns true when 1st argument _is after_ 2nd one.
There is one possible problem with this poll loop: the scheduler could put the thread to sleep, and it does not get woken up for OCELOT_FDMA_CH_SAFE_TIMEOUT_US. During that time, the hardware has done its thing, but you exit the while loop and return -ETIMEDOUT.
Fix it by using sane poll API that avoids all problems described above
--- drivers/net/ethernet/mscc/ocelot_fdma.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-)
drivers/net/ethernet/mscc/ocelot_fdma.c +8 -9
@@ -94,19 +94,18 @@ static void ocelot_fdma_activate_chan(struct ocelot *ocelot, dma_addr_t dma,ocelot_fdma_writel (ocelot,MSCC_FDMA_CH_ACTIVATE ,BIT (chan)); } +static u32 ocelot_fdma_read_ch_safe (struct ocelot *ocelot) +{ +return ocelot_fdma_readl (ocelot,MSCC_FDMA_CH_SAFE ); +} +static int ocelot_fdma_wait_chan_safe (struct ocelot *ocelot,int chan) { -unsigned long timeout;u32 safe; - timeout = jiffies +usecs_to_jiffies (OCELOT_FDMA_CH_SAFE_TIMEOUT_US ); -do { - safe =ocelot_fdma_readl (ocelot,MSCC_FDMA_CH_SAFE ); -if (safe &BIT (chan)) -return 0 ; - }while (time_after (jiffies, timeout)); - -return -ETIMEDOUT ; +return readx_poll_timeout_atomic (ocelot_fdma_read_ch_safe, ocelot, safe, + safe &BIT (chan),0 , +OCELOT_FDMA_CH_SAFE_TIMEOUT_US ); } static void ocelot_fdma_dcb_set_data(struct ocelot_fdma_dcb *dcb,
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 69/231 sysctl: Fix data races in proc_dointvec().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 1f1be04b4d48a2475ea1aab46a99221bfc5c0968 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_dointvec() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_dointvec() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
kernel/sysctl.c +3 -3
@@ -518,14 +518,14 @@ static int do_proc_dointvec_conv(bool *negp, unsigned long *lvalp,if (*negp) {if (*lvalp > (unsigned long )INT_MAX +1 )return -EINVAL ; - *valp = -*lvalp; +WRITE_ONCE (*valp, -*lvalp); }else {if (*lvalp > (unsigned long )INT_MAX )return -EINVAL ; - *valp = *lvalp; +WRITE_ONCE (*valp, *lvalp); } }else { -int val = *valp; +int val =READ_ONCE (*valp);if (val <0 ) { *negp = true; *lvalp = -(unsigned long )val;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 70/231 sysctl: Fix data races in proc_douintvec().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 4762b532ec9539755aab61445d5da6e1926ccb99 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_douintvec() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_douintvec() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
kernel/sysctl.c +2 -2
@@ -544,9 +544,9 @@ static int do_proc_douintvec_conv(unsigned long *lvalp,if (write) {if (*lvalp >UINT_MAX )return -EINVAL ; - *valp = *lvalp; +WRITE_ONCE (*valp, *lvalp); }else { -unsigned int val = *valp; +unsigned int val =READ_ONCE (*valp); *lvalp = (unsigned long )val; }return 0 ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 71/231 sysctl: Fix data races in proc_dointvec_minmax().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit f613d86d014b6375a4085901de39406598121e35 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_dointvec_minmax() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_dointvec_minmax() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/sysctl.c +1 -1
@@ -929,7 +929,7 @@ static int do_proc_dointvec_minmax_conv(bool *negp, unsigned long *lvalp,if ((param->min && *param->min > tmp) || (param->max && *param->max < tmp))return -EINVAL ; - *valp = tmp; +WRITE_ONCE (*valp, tmp); }return 0 ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 72/231 sysctl: Fix data races in proc_douintvec_minmax().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 2d3b559df3ed39258737789aae2ae7973d205bc1 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_douintvec_minmax() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_douintvec_minmax() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/sysctl.c +1 -1
@@ -995,7 +995,7 @@ static int do_proc_douintvec_minmax_conv(unsigned long *lvalp, (param->max && *param->max < tmp))return -ERANGE ; - *valp = tmp; +WRITE_ONCE (*valp, tmp); }return 0 ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 73/231 sysctl: Fix data races in proc_doulongvec_minmax().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit c31bcc8fb89fc2812663900589c6325ba35d9a65 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_doulongvec_minmax() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_doulongvec_minmax() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
kernel/sysctl.c +2 -2
@@ -1162,9 +1162,9 @@ static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, err = -EINVAL ;break ; } - *i = val; +WRITE_ONCE (*i, val); }else { - val = convdiv * (*i) / convmul; + val = convdiv *READ_ONCE (*i) / convmul;if (!first)proc_put_char (&buffer, &left,'\t' );proc_put_long (&buffer, &left, val, false);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 74/231 sysctl: Fix data races in proc_dointvec_jiffies().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit e877820877663fbae8cb9582ea597a7230b94df3 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_dointvec_jiffies() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_dointvec_jiffies() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
kernel/sysctl.c +5 -2
@@ -1245,9 +1245,12 @@ static int do_proc_dointvec_jiffies_conv(bool *negp, unsigned long *lvalp,if (write) {if (*lvalp >INT_MAX /HZ )return 1 ; - *valp = *negp ? -(*lvalp*HZ ) : (*lvalp*HZ ); +if (*negp) +WRITE_ONCE (*valp, -*lvalp *HZ ); +else +WRITE_ONCE (*valp, *lvalp *HZ ); }else { -int val = *valp; +int val =READ_ONCE (*valp);unsigned long lval;if (val <0 ) { *negp = true;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 75/231 tcp: Fix a data-race around sysctl_tcp_max_orphans.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 47e6ab24e8c6e3ca10ceb5835413f401f90de4bf ]
While reading sysctl_tcp_max_orphans, it can be changed concurrently. So, we need to add READ_ONCE() to avoid a data-race.
--- net/ipv4/tcp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/ipv4/tcp.c +2 -1
@@ -2735,7 +2735,8 @@ static void tcp_orphan_update(struct timer_list *unused)static bool tcp_too_many_orphans (int shift) { -return READ_ONCE (tcp_orphan_cache) << shift > sysctl_tcp_max_orphans; +return READ_ONCE (tcp_orphan_cache) << shift > +READ_ONCE (sysctl_tcp_max_orphans); }bool tcp_check_oom (struct sock *sk,int shift)
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 76/231 inetpeer: Fix data-races around sysctl.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 3d32edf1f3c38d3301f6434e56316f293466d7fb ]
While reading inetpeer sysctl variables, they can be changed concurrently. So, we need to add READ_ONCE() to avoid data-races.
--- net/ipv4/inetpeer.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-)
net/ipv4/inetpeer.c +8 -4
@@ -141,16 +141,20 @@ static void inet_peer_gc(struct inet_peer_base *base,struct inet_peer *gc_stack[], unsigned int gc_cnt) { + int peer_threshold, peer_maxttl, peer_minttl;struct inet_peer *p;__u32 delta, ttl;int i; -if (base->total >= inet_peer_threshold) + peer_threshold =READ_ONCE (inet_peer_threshold); + peer_maxttl =READ_ONCE (inet_peer_maxttl); + peer_minttl =READ_ONCE (inet_peer_minttl); + +if (base->total >= peer_threshold) ttl =0 ;/* be aggressive */ else - ttl = inet_peer_maxttl - - (inet_peer_maxttl - inet_peer_minttl) /HZ * - base->total / inet_peer_threshold *HZ ; + ttl = peer_maxttl - (peer_maxttl - peer_minttl) /HZ * + base->total / peer_threshold *HZ ;for (i =0 ; i < gc_cnt; i++) { p = gc_stack[i];
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 77/231 net: Fix data-races around sysctl_mem.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 310731e2f1611d1d13aae237abcf8e66d33345d5 ]
While reading .sysctl_mem, it can be changed concurrently. So, we need to add READ_ONCE() to avoid data-races.
--- include/net/sock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/net/sock.h +1 -1
@@ -1534,7 +1534,7 @@ void __sk_mem_reclaim(struct sock *sk, int amount);/* sysctl_mem values are in pages, we convert them in SK_MEM_QUANTUM units */ static inline long sk_prot_mem_limits (const struct sock *sk,int index) { -long val = sk->sk_prot->sysctl_mem[index]; +long val =READ_ONCE (sk->sk_prot->sysctl_mem[index]);#if PAGE_SIZE >SK_MEM_QUANTUM val <<=PAGE_SHIFT -SK_MEM_QUANTUM_SHIFT ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 78/231 cipso: Fix data-races around sysctl.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit dd44f04b9214adb68ef5684ae87a81ba03632250 ]
While reading cipso sysctl variables, they can be changed concurrently. So, we need to add READ_ONCE() to avoid data-races.
--- Documentation/networking/ip-sysctl.rst | 2 +- net/ipv4/cipso_ipv4.c | 12 +++++++----- 2 files changed, 8 insertions(+), 6 deletions(-)
Documentation/networking/ip-sysctl.rst +1 -1
@@ -1085,7 +1085,7 @@ cipso_cache_enable - BOOLEAN cipso_cache_bucket_size - INTEGER The CIPSO label cache consists of a fixed size hash table with each hash bucket containing a number of cache entries. This variable limits - the number of entries in each hash bucket; the larger the value the + the number of entries in each hash bucket; the larger the value is, the more CIPSO label mappings that can be cached. When the number of entries in a given hash bucket reaches this limit adding new entries causes the oldest entry in the bucket to be removed to make room.
net/ipv4/cipso_ipv4.c +7 -5
@@ -239,7 +239,7 @@ static int cipso_v4_cache_check(const unsigned char *key,struct cipso_v4_map_cache_entry *prev_entry =NULL ;u32 hash; -if (!cipso_v4_cache_enabled) +if (!READ_ONCE (cipso_v4_cache_enabled))return -ENOENT ; hash =cipso_v4_map_cache_hash (key, key_len);
@@ -296,13 +296,14 @@ static int cipso_v4_cache_check(const unsigned char *key,int cipso_v4_cache_add (const unsigned char *cipso_ptr,const struct netlbl_lsm_secattr *secattr) { +int bkt_size =READ_ONCE (cipso_v4_cache_bucketsize);int ret_val = -EPERM ;u32 bkt;struct cipso_v4_map_cache_entry *entry =NULL ;struct cipso_v4_map_cache_entry *old_entry =NULL ;u32 cipso_ptr_len; -if (!cipso_v4_cache_enabled || cipso_v4_cache_bucketsize <=0 ) +if (!READ_ONCE (cipso_v4_cache_enabled) || bkt_size <=0 )return 0 ; cipso_ptr_len = cipso_ptr[1 ];
@@ -322,7 +323,7 @@ int cipso_v4_cache_add(const unsigned char *cipso_ptr, bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETS -1 );spin_lock_bh (&cipso_v4_cache[bkt].lock); -if (cipso_v4_cache[bkt].size < cipso_v4_cache_bucketsize) { +if (cipso_v4_cache[bkt].size < bkt_size) {list_add (&entry->list, &cipso_v4_cache[bkt].list); cipso_v4_cache[bkt].size +=1 ; }else {
@@ -1199,7 +1200,8 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def,/* This will send packets using the "optimized" format when * possible as specified in section 3.4.2.6 of the * CIPSO draft. */ -if (cipso_v4_rbm_optfmt && ret_val >0 && ret_val <=10 ) +if (READ_ONCE (cipso_v4_rbm_optfmt) && ret_val >0 && + ret_val <=10 ) tag_len =14 ;else tag_len =4 + ret_val;
@@ -1603,7 +1605,7 @@ int cipso_v4_validate(const struct sk_buff *skb, unsigned char **option)* all the CIPSO validations here but it doesn't * really specify _exactly_ what we need to validate * ... so, just make it a sysctl tunable. */ -if (cipso_v4_rbm_strictvalid) { +if (READ_ONCE (cipso_v4_rbm_strictvalid)) {if (cipso_v4_map_lvl_valid (doi_def, tag[3 ]) <0 ) { err_offset = opt_iter +3 ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 79/231 icmp: Fix data-races around sysctl.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 48d7ee321ea5182c6a70782aa186422a70e67e22 ]
While reading icmp sysctl variables, they can be changed concurrently. So, we need to add READ_ONCE() to avoid data-races.
--- net/ipv4/icmp.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv4/icmp.c +3 -2
@@ -253,11 +253,12 @@ bool icmp_global_allow(void)spin_lock (&icmp_global.lock); delta =min_t (u32, now - icmp_global.stamp,HZ );if (delta >=HZ /50 ) { - incr = sysctl_icmp_msgs_per_sec * delta /HZ ; + incr =READ_ONCE (sysctl_icmp_msgs_per_sec) * delta /HZ ;if (incr)WRITE_ONCE (icmp_global.stamp, now); } - credit =min_t (u32, icmp_global.credit + incr, sysctl_icmp_msgs_burst); + credit =min_t (u32, icmp_global.credit + incr, +READ_ONCE (sysctl_icmp_msgs_burst));if (credit) {/* We want to use a credit of one in average, but need to randomize * it for security reasons.
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 80/231 ipv4: Fix a data-race around sysctl_fib_sync_mem.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 73318c4b7dbd0e781aaababff17376b2894745c0 ]
While reading sysctl_fib_sync_mem, it can be changed concurrently. So, we need to add READ_ONCE() to avoid a data-race.
--- net/ipv4/fib_trie.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/fib_trie.c +1 -1
@@ -498,7 +498,7 @@ static void tnode_free(struct key_vector *tn) tn = container_of(head, struct tnode, rcu)->kv; } -if (tnode_free_size >= sysctl_fib_sync_mem) { +if (tnode_free_size >=READ_ONCE (sysctl_fib_sync_mem)) { tnode_free_size =0 ;synchronize_rcu (); }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 81/231 ARM: dts: at91: sama5d2: Fix typo in i2s1 node
In reply to Greg Kroah-HartmanFrom: Ryan Wanner <[email protected]>
[ Upstream commit 2fdf15b50a46e366740df4cccbe2343269b4ff55 ]
Fix typo in i2s1 causing errors in dt binding validation. Change assigned-parrents to assigned-clock-parents to match i2s0 node formatting.
[claudiu.beznea: use imperative addressing in commit description, remove
blank line after fixes tag, fix typo in commit message]
--- arch/arm/boot/dts/sama5d2.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/sama5d2.dtsi +1 -1
@@ -1124,7 +1124,7 @@ AT91_XDMAC_DT_PERID(33))>, clocks = <&pmc PMC_TYPE_PERIPHERAL 55>, <&pmc PMC_TYPE_GCK 55>; clock-names = "pclk", "gclk"; assigned-clocks = <&pmc PMC_TYPE_CORE PMC_I2S1_MUX>; - assigned-parrents = <&pmc PMC_TYPE_GCK 55>; + assigned-clock-parents = <&pmc PMC_TYPE_GCK 55>; status = "disabled"; };
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 82/231 ARM: dts: sunxi: Fix SPI NOR campatible on Orange Pi Zero
In reply to Greg Kroah-HartmanFrom: Michal Suchanek <[email protected]>
[ Upstream commit 884b66976a7279ee889ba885fe364244d50b79e7 ]
The device tree should include generic "jedec,spi-nor" compatible, and a manufacturer-specific one.
The macronix part is what is shipped on the boards that come with a flash chip.
--- arch/arm/boot/dts/sun8i-h2-plus-orangepi-zero.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/sun8i-h2-plus-orangepi-zero.dts +1 -1
@@ -169,7 +169,7 @@ &spi0 { flash@0 { #address-cells = <1>; #size-cells = <1>; - compatible = "mxicy,mx25l1606e", "winbond,w25q128"; + compatible = "mxicy,mx25l1606e", "jedec,spi-nor"; reg = <0>; spi-max-frequency = <40000000>; };
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 83/231 arm64: dts: broadcom: bcm4908: Fix timer node for BCM4906 SoC
In reply to Greg Kroah-HartmanFrom: William Zhang <[email protected]>
[ Upstream commit b4a544e415e9be33b37d9bfa9d9f9f4d13f553d6 ]
The cpu mask value in interrupt property inherits from bcm4908.dtsi which sets to four cpus. Correct the value to two cpus for dual core BCM4906 SoC.
--- arch/arm64/boot/dts/broadcom/bcm4908/bcm4906.dtsi | 8 ++++++++ 1 file changed, 8 insertions(+)
arch/arm64/boot/dts/broadcom/bcm4908/bcm4906.dtsi +8 -0
@@ -9,6 +9,14 @@ cpus { /delete-node/ cpu@3; }; + timer { + compatible = "arm,armv8-timer"; + interrupts = <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(2) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(2) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(2) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(2) | IRQ_TYPE_LEVEL_LOW)>; + }; + pmu { compatible = "arm,cortex-a53-pmu"; interrupts = <GIC_SPI 9 IRQ_TYPE_LEVEL_HIGH>,
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 84/231 arm64: dts: broadcom: bcm4908: Fix cpu node for smp boot
In reply to Greg Kroah-HartmanFrom: William Zhang <[email protected]>
[ Upstream commit 8bd582ae9a71d7f14c4e0c735b2eacaf7516d626 ]
Add spin-table enable-method and cpu-release-addr properties for cpu0 node. This is required by all ARMv8 SoC. Otherwise some bootloader like u-boot can not update cpu-release-addr and linux fails to start up secondary cpus.
--- arch/arm64/boot/dts/broadcom/bcm4908/bcm4908.dtsi | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/boot/dts/broadcom/bcm4908/bcm4908.dtsi +2 -0
@@ -29,6 +29,8 @@ cpu0: cpu@0 { device_type = "cpu"; compatible = "brcm,brahma-b53"; reg = <0x0>; + enable-method = "spin-table"; + cpu-release-addr = <0x0 0xfff8>; next-level-cache = <&l2>; };
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 85/231 netfilter: nf_log: incorrect offset to network header
In reply to Greg Kroah-HartmanFrom: Pablo Neira Ayuso <[email protected]>
[ Upstream commit 7a847c00eeba9744353ecdfad253143b9115678a ]
NFPROTO_ARP is expecting to find the ARP header at the network offset.
In the particular case of ARP, HTYPE= field shows the initial bytes of the ethernet header destination MAC address.
netdev out: IN= OUT=bridge0 MACSRC=c2:76:e5:71:e1:de MACDST=36:b0:4a:e2:72:ea MACPROTO=0806 ARP HTYPE=14000 PTYPE=0x4ae2 OPCODE=49782
NFPROTO_NETDEV egress hook is also expecting to find the IP headers at the network offset.
--- net/netfilter/nf_log_syslog.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
net/netfilter/nf_log_syslog.c +4 -4
@@ -61,7 +61,7 @@ dump_arp_packet(struct nf_log_buf *m,unsigned int logflags;struct arphdr _arph; - ah =skb_header_pointer (skb,0 ,sizeof (_arph), &_arph); + ah =skb_header_pointer (skb, nhoff,sizeof (_arph), &_arph);if (!ah) {nf_log_buf_add (m,"TRUNCATED" );return ;
@@ -90,7 +90,7 @@ dump_arp_packet(struct nf_log_buf *m, ah->ar_pln != sizeof(__be32))return ; - ap =skb_header_pointer (skb,sizeof (_arph),sizeof (_arpp), &_arpp); + ap =skb_header_pointer (skb, nhoff +sizeof (_arph),sizeof (_arpp), &_arpp);if (!ap) {nf_log_buf_add (m," INCOMPLETE [%zu bytes]" , skb->len -sizeof (_arph));
@@ -144,7 +144,7 @@ static void nf_log_arp_packet(struct net *net, u_int8_t pf,nf_log_dump_packet_common (m, pf, hooknum, skb, in, out, loginfo, prefix); -dump_arp_packet (m, loginfo, skb,0 ); +dump_arp_packet (m, loginfo, skb,skb_network_offset (skb));nf_log_buf_close (m); }
@@ -829,7 +829,7 @@ static void nf_log_ip_packet(struct net *net, u_int8_t pf,if (in)dump_ipv4_mac_header (m, loginfo, skb); -dump_ipv4_packet (net, m, loginfo, skb,0 ); +dump_ipv4_packet (net, m, loginfo, skb,skb_network_offset (skb));nf_log_buf_close (m); }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 86/231 nfp: fix issue of skb segments exceeds descriptor limitation
In reply to Greg Kroah-HartmanFrom: Baowen Zheng <[email protected]>
[ Upstream commit 9c840d5f9aaef87e65db900bae21c70b059aba5f ]
TCP packets will be dropped if the segments number in the tx skb exceeds limitation when sending iperf3 traffic with --zerocopy option.
we make the following changes:
Get nr_frags in nfp_nfdk_tx_maybe_close_block instead of passing from outside because it will be changed after skb_linearize operation.
Fill maximum dma_len in first tx descriptor to make sure the whole head is included in the first descriptor.
--- drivers/net/ethernet/netronome/nfp/nfdk/dp.c | 33 +++++++++++++++----- 1 file changed, 25 insertions(+), 8 deletions(-)
drivers/net/ethernet/netronome/nfp/nfdk/dp.c +25 -8
@@ -125,17 +125,18 @@ nfp_nfdk_tx_csum(struct nfp_net_dp *dp, struct nfp_net_r_vector *r_vec,static int nfp_nfdk_tx_maybe_close_block (struct nfp_net_tx_ring *tx_ring, -unsigned int nr_frags,struct sk_buff *skb) +struct sk_buff *skb) {unsigned int n_descs, wr_p, nop_slots;const skb_frag_t *frag, *fend;struct nfp_nfdk_tx_desc *txd; +unsigned int nr_frags;unsigned int wr_idx;int err; recount_descs: n_descs =nfp_nfdk_headlen_to_segs (skb_headlen (skb)); - + nr_frags =skb_shinfo (skb)->nr_frags; frag =skb_shinfo (skb)->frags; fend = frag + nr_frags;for (; frag < fend; frag++)
@@ -281,10 +282,13 @@ netdev_tx_t nfp_nfdk_tx(struct sk_buff *skb, struct net_device *netdev)if (unlikely ((int )metadata <0 )) goto err_flush; - nr_frags =skb_shinfo (skb)->nr_frags; -if (nfp_nfdk_tx_maybe_close_block (tx_ring, nr_frags, skb)) +if (nfp_nfdk_tx_maybe_close_block (tx_ring, skb)) goto err_flush; +/* nr_frags will change after skb_linearize so we get nr_frags after +* nfp_nfdk_tx_maybe_close_block function +*/ + nr_frags =skb_shinfo (skb)->nr_frags;/* DMA map all */ wr_idx =D_IDX (tx_ring, tx_ring->wr_p); txd = &tx_ring->ktxds[wr_idx];
@@ -310,7 +314,16 @@ netdev_tx_t nfp_nfdk_tx(struct sk_buff *skb, struct net_device *netdev)/* FIELD_PREP() implicitly truncates to chunk */ dma_len -=1 ; - dlen_type =FIELD_PREP (NFDK_DESC_TX_DMA_LEN_HEAD , dma_len) | + +/* We will do our best to pass as much data as we can in descriptor +* and we need to make sure the first descriptor includes whole head +* since there is limitation in firmware side. Sometimes the value of +* dma_len bitwise and NFDK_DESC_TX_DMA_LEN_HEAD will less than +* headlen. +*/ + dlen_type =FIELD_PREP (NFDK_DESC_TX_DMA_LEN_HEAD , + dma_len >NFDK_DESC_TX_DMA_LEN_HEAD ? +NFDK_DESC_TX_DMA_LEN_HEAD : dma_len) |FIELD_PREP (NFDK_DESC_TX_TYPE_HEAD , type); txd->dma_len_type =cpu_to_le16 (dlen_type);
@@ -925,7 +938,9 @@ nfp_nfdk_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,/* FIELD_PREP() implicitly truncates to chunk */ dma_len -=1 ; - dlen_type =FIELD_PREP (NFDK_DESC_TX_DMA_LEN_HEAD , dma_len) | + dlen_type =FIELD_PREP (NFDK_DESC_TX_DMA_LEN_HEAD , + dma_len >NFDK_DESC_TX_DMA_LEN_HEAD ? +NFDK_DESC_TX_DMA_LEN_HEAD : dma_len) |FIELD_PREP (NFDK_DESC_TX_TYPE_HEAD , type); txd->dma_len_type =cpu_to_le16 (dlen_type);
@@ -1303,7 +1318,7 @@ nfp_nfdk_ctrl_tx_one(struct nfp_net *nn, struct nfp_net_r_vector *r_vec, skb_push(skb, 4)); } -if (nfp_nfdk_tx_maybe_close_block (tx_ring,0 , skb)) +if (nfp_nfdk_tx_maybe_close_block (tx_ring, skb)) goto err_free;/* DMA map all */
@@ -1328,7 +1343,9 @@ nfp_nfdk_ctrl_tx_one(struct nfp_net *nn, struct nfp_net_r_vector *r_vec, txbuf++; dma_len -=1 ; - dlen_type =FIELD_PREP (NFDK_DESC_TX_DMA_LEN_HEAD , dma_len) | + dlen_type =FIELD_PREP (NFDK_DESC_TX_DMA_LEN_HEAD , + dma_len >NFDK_DESC_TX_DMA_LEN_HEAD ? +NFDK_DESC_TX_DMA_LEN_HEAD : dma_len) |FIELD_PREP (NFDK_DESC_TX_TYPE_HEAD , type); txd->dma_len_type =cpu_to_le16 (dlen_type);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 87/231 vlan: fix memory leak in vlan_newlink()
In reply to Greg Kroah-HartmanFrom: Eric Dumazet <[email protected]>
[ Upstream commit 72a0b329114b1caa8e69dfa7cdad1dd3c69b8602 ]
Blamed commit added back a bug I fixed in commit 9bbd917e0bec ("vlan: fix memory leak in vlan_dev_set_egress_priority")
If a memory allocation fails in vlan_changelink() after other allocations succeeded, we need to call vlan_dev_free_egress_priority()
to free all allocated memory because after a failed ->newlink() we do not call any methods like ndo_uninit() or dev->priv_destructor().
In following example, if the allocation for last element 2000:2001 fails, we need to free eight prior allocations:
ip link add link dummy0 dummy0.100 type vlan id 100 \
egress-qos-map 1:2 2:3 3:4 4:5 5:6 6:7 7:8 8:9 2000:2001
syzbot report was:
BUG: memory leak
unreferenced object 0xffff888117bd1060 (size 32):
comm "syz-executor408", pid 3759, jiffies 4294956555 (age 34.090s) hex dump (first 32 bytes):
09 00 00 00 00 a0 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace:
[<ffffffff83fc60ad>] kmalloc include/linux/slab.h:600 [inline] [<ffffffff83fc60ad>] vlan_dev_set_egress_priority+0xed/0x170 net/8021q/vlan_dev.c:193 [<ffffffff83fc6628>] vlan_changelink+0x178/0x1d0 net/8021q/vlan_netlink.c:128 [<ffffffff83fc67c8>] vlan_newlink+0x148/0x260 net/8021q/vlan_netlink.c:185 [<ffffffff838b1278>] rtnl_newlink_create net/core/rtnetlink.c:3363 [inline] [<ffffffff838b1278>] __rtnl_newlink+0xa58/0xdc0 net/core/rtnetlink.c:3580 [<ffffffff838b1629>] rtnl_newlink+0x49/0x70 net/core/rtnetlink.c:3593 [<ffffffff838ac66c>] rtnetlink_rcv_msg+0x21c/0x5c0 net/core/rtnetlink.c:6089 [<ffffffff839f9c37>] netlink_rcv_skb+0x87/0x1d0 net/netlink/af_netlink.c:2501 [<ffffffff839f8da7>] netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline] [<ffffffff839f8da7>] netlink_unicast+0x397/0x4c0 net/netlink/af_netlink.c:1345 [<ffffffff839f9266>] netlink_sendmsg+0x396/0x710 net/netlink/af_netlink.c:1921 [<ffffffff8384dbf6>] sock_sendmsg_nosec net/socket.c:714 [inline] [<ffffffff8384dbf6>] sock_sendmsg+0x56/0x80 net/socket.c:734 [<ffffffff8384e15c>] ____sys_sendmsg+0x36c/0x390 net/socket.c:2488 [<ffffffff838523cb>] ___sys_sendmsg+0x8b/0xd0 net/socket.c:2542 [<ffffffff838525b8>] __sys_sendmsg net/socket.c:2571 [inline] [<ffffffff838525b8>] __do_sys_sendmsg net/socket.c:2580 [inline] [<ffffffff838525b8>] __se_sys_sendmsg net/socket.c:2578 [inline] [<ffffffff838525b8>] __x64_sys_sendmsg+0x78/0xf0 net/socket.c:2578 [<ffffffff845ad8d5>] do_syscall_x64 arch/x86/entry/common.c:50 [inline] [<ffffffff845ad8d5>] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 [<ffffffff8460006a>] entry_SYSCALL_64_after_hwframe+0x46/0xb0
--- net/8021q/vlan_netlink.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
net/8021q/vlan_netlink.c +7 -3
@@ -182,10 +182,14 @@ static int vlan_newlink(struct net *src_net, struct net_device *dev,else if (dev->mtu > max_mtu)return -EINVAL ; +/* Note: If this initial vlan_changelink() fails, we need +* to call vlan_dev_free_egress_priority() to free memory. +*/ err =vlan_changelink (dev, tb, data, extack); -if (err) -return err; - err =register_vlan_dev (dev, extack); + +if (!err) + err =register_vlan_dev (dev, extack); +if (err)vlan_dev_free_egress_priority (dev);return err;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 88/231 netfilter: nf_tables: replace BUG_ON by element length check
In reply to Greg Kroah-HartmanFrom: Pablo Neira Ayuso <[email protected]>
[ Upstream commit c39ba4de6b0a843bec5d46c2b6f2064428dada5e ]
BUG_ON can be triggered from userspace with an element with a large userdata area. Replace it by length check and return EINVAL instead. Over time extensions have been growing in size.
Pick a sufficiently old Fixes: tag to propagate this fix.
--- include/net/netfilter/nf_tables.h | 14 ++++--- net/netfilter/nf_tables_api.c | 72 ++++++++++++++++++++++++++------------ 2 files changed, 60 insertions(+), 26 deletions(-)
include/net/netfilter/nf_tables.h +9 -5
@@ -657,18 +657,22 @@ static inline void nft_set_ext_prepare(s tmpl->len =sizeof (struct nft_set_ext ); } -static inline void nft_set_ext_add_length (struct nft_set_ext_tmpl *tmpl,u8 id, -unsigned int len) +static inline int nft_set_ext_add_length (struct nft_set_ext_tmpl *tmpl,u8 id, +unsigned int len) { tmpl->len =ALIGN (tmpl->len, nft_set_ext_types[id].align); -BUG_ON (tmpl->len >U8_MAX ); +if (tmpl->len >U8_MAX ) +return -EINVAL ; + tmpl->offset[id] = tmpl->len; tmpl->len += nft_set_ext_types[id].len + len; + +return 0 ; } -static inline void nft_set_ext_add (struct nft_set_ext_tmpl *tmpl,u8 id) +static inline int nft_set_ext_add (struct nft_set_ext_tmpl *tmpl,u8 id) { -nft_set_ext_add_length (tmpl, id,0 ); +return nft_set_ext_add_length (tmpl, id,0 ); } static inline void nft_set_ext_init(struct nft_set_ext *ext,
net/netfilter/nf_tables_api.c +51 -21
148 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 89/231 RISC-V: KVM: Fix SRCU deadlock caused by kvm_riscv_check_vcpu_requests()
In reply to Greg Kroah-HartmanFrom: Anup Patel <[email protected]>
[ Upstream commit be82abe6a76ba8e76f25312566182b0f13c4fbf9 ]
The kvm_riscv_check_vcpu_requests() is called with SRCU read lock held and for KVM_REQ_SLEEP request it will block the VCPU without releasing SRCU read lock. This causes KVM ioctls (such as KVM_IOEVENTFD) from other VCPUs of the same Guest/VM to hang/deadlock if there is any synchronize_srcu() or synchronize_srcu_expedited() in the path.
To fix the above in kvm_riscv_check_vcpu_requests(), we should do SRCU read unlock before blocking the VCPU and do SRCU read lock after VCPU wakeup.
--- arch/riscv/kvm/vcpu.c | 2 ++ 1 file changed, 2 insertions(+)
arch/riscv/kvm/vcpu.c +2 -0
@@ -673,9 +673,11 @@ static void kvm_riscv_check_vcpu_requests(struct kvm_vcpu *vcpu)if (kvm_request_pending (vcpu)) {if (kvm_check_request (KVM_REQ_SLEEP , vcpu)) { +kvm_vcpu_srcu_read_unlock (vcpu);rcuwait_wait_event (wait, (!vcpu->arch.power_off) && (!vcpu->arch.pause),TASK_INTERRUPTIBLE ); +kvm_vcpu_srcu_read_lock (vcpu);if (vcpu->arch.power_off || vcpu->arch.pause) {/*
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 90/231 drm/i915/gvt: IS_ERR() vs NULL bug in intel_gvt_update_reg_whitelist()
In reply to Greg Kroah-HartmanFrom: Dan Carpenter <[email protected]>
[ Upstream commit e87197fbd137c888fd6c871c72fe7e89445dd015 ]
The shmem_pin_map() function returns NULL, it doesn't return error pointers.
--- drivers/gpu/drm/i915/gvt/cmd_parser.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/gvt/cmd_parser.c +3 -3
@@ -3117,9 +3117,9 @@ void intel_gvt_update_reg_whitelist(struct intel_vgpu *vgpu)continue ; vaddr =shmem_pin_map (engine->default_state); -if (IS_ERR (vaddr)) { -gvt_err ("failed to map %s->default state, err:%zd\n" , - engine->name,PTR_ERR (vaddr)); +if (!vaddr) { +gvt_err ("failed to map %s->default state\n" , + engine->name);return ; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 91/231 xen/gntdev: Ignore failure to unmap INVALID_GRANT_HANDLE
In reply to Greg Kroah-HartmanFrom: Demi Marie Obenour <[email protected]>
[ Upstream commit 166d3863231667c4f64dee72b77d1102cdfad11f ]
The error paths of gntdev_mmap() can call unmap_grant_pages() even though not all of the pages have been successfully mapped. This will trigger the WARN_ON()s in __unmap_grant_pages_done(). The number of warnings can be very large; I have observed thousands of lines of warnings in the systemd journal.
Avoid this problem by only warning on unmapping failure if the handle being unmapped is not INVALID_GRANT_HANDLE. The handle field of any page that was not successfully mapped will be INVALID_GRANT_HANDLE, so this catches all cases where unmapping can legitimately fail.
--- drivers/xen/gntdev.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/xen/gntdev.c +4 -2
@@ -396,13 +396,15 @@ static void __unmap_grant_pages_done(int result,unsigned int offset = data->unmap_ops - map->unmap_ops;for (i =0 ; i < data->count; i++) { -WARN_ON (map->unmap_ops[offset+i].status); +WARN_ON (map->unmap_ops[offset + i].status != GNTST_okay && + map->unmap_ops[offset + i].handle !=INVALID_GRANT_HANDLE );pr_debug ("unmap handle=%d st=%d\n" , map->unmap_ops[offset+i].handle, map->unmap_ops[offset+i].status); map->unmap_ops[offset+i].handle =INVALID_GRANT_HANDLE ;if (use_ptemod) { -WARN_ON (map->kunmap_ops[offset+i].status); +WARN_ON (map->kunmap_ops[offset + i].status != GNTST_okay && + map->kunmap_ops[offset + i].handle !=INVALID_GRANT_HANDLE );pr_debug ("kunmap handle=%u st=%d\n" , map->kunmap_ops[offset+i].handle, map->kunmap_ops[offset+i].status);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 92/231 mptcp: fix subflow traversal at disconnect time
In reply to Greg Kroah-HartmanFrom: Paolo Abeni <[email protected]>
[ Upstream commit 5c835bb142d4013c2ab24bff5ae9f6709a39cbcf ]
At disconnect time the MPTCP protocol traverse the subflows list closing each of them. In some circumstances - MPJ subflow, passive MPTCP socket, the latter operation can remove the subflow from the list, invalidating the current iterator.
Address the issue using the safe list traversing helper variant.
--- net/mptcp/protocol.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/mptcp/protocol.c +2 -2
@@ -2840,12 +2840,12 @@ static void mptcp_copy_inaddrs(struct sostatic int mptcp_disconnect (struct sock *sk,int flags) { -struct mptcp_subflow_context *subflow; +struct mptcp_subflow_context *subflow, *tmp;struct mptcp_sock *msk =mptcp_sk (sk);inet_sk_state_store (sk,TCP_CLOSE ); -mptcp_for_each_subflow (msk, subflow) { +list_for_each_entry_safe (subflow, tmp, &msk->conn_list, node) {struct sock *ssk =mptcp_subflow_tcp_sock (subflow);__mptcp_close_ssk (sk, ssk, subflow,MPTCP_CF_FASTCLOSE );
Greg Kroah-Hartman
maintainerPATCH 93/231 NFSD: Decode NFSv4 birth time attribute
In reply to Greg Kroah-HartmanFrom: Chuck Lever <[email protected]>
[ Upstream commit 5b2f3e0777da2a5dd62824bbe2fdab1d12caaf8f ]
NFSD has advertised support for the NFSv4 time_create attribute since commit e377a3e698fb ("nfsd: Add support for the birth time attribute").
Igor Mammedov reports that Mac OS clients attempt to set the NFSv4 birth time attribute via OPEN(CREATE) and SETATTR if the server indicates that it supports it, but since the above commit was merged, those attempts now fail.
Table 5 in RFC 8881 lists the time_create attribute as one that can be both set and retrieved, but the above commit did not add server support for clients to provide a time_create attribute. IMO that's a bug in our implementation of the NFSv4 protocol, which this commit addresses.
Whether NFSD silently ignores the new birth time or actually sets it is another matter. I haven't found another filesystem service in the Linux kernel that enables users or clients to modify a file's birth time attribute.
This commit reflects my (perhaps incorrect) understanding of whether Linux users can set a file's birth time. NFSD will now recognize a time_create attribute but it ignores its value. It clears the time_create bit in the returned attribute bitmask to indicate that the value was not used.
--- fs/nfsd/nfs4xdr.c | 9 +++++++++ fs/nfsd/nfsd.h | 3 ++- 2 files changed, 11 insertions(+), 1 deletion(-)
fs/nfsd/nfs4xdr.c +9 -0
@@ -470,6 +470,15 @@ nfsd4_decode_fattr4(struct nfsd4_compoundargs *argp, u32 *bmval, u32 bmlen,return nfserr_bad_xdr; } } +if (bmval[1 ] &FATTR4_WORD1_TIME_CREATE ) { +struct timespec64 ts; + +/* No Linux filesystem supports setting this attribute. */ + bmval[1 ] &= ~FATTR4_WORD1_TIME_CREATE ; + status =nfsd4_decode_nfstime4 (argp, &ts); +if (status) +return status; + }if (bmval[1 ] &FATTR4_WORD1_TIME_MODIFY_SET ) {u32 set_it;
fs/nfsd/nfsd.h +2 -1
@@ -460,7 +460,8 @@ static inline bool nfsd_attrs_supported(u32 minorversion, const u32 *bmval) (FATTR4_WORD0_SIZE | FATTR4_WORD0_ACL)#define NFSD_WRITEABLE_ATTRS_WORD1 \ (FATTR4_WORD1_MODE | FATTR4_WORD1_OWNER | FATTR4_WORD1_OWNER_GROUP \ - | FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET) + | FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_CREATE \ + | FATTR4_WORD1_TIME_MODIFY_SET)#ifdef CONFIG_NFSD_V4_SECURITY_LABEL #define MAYBE_FATTR4_WORD2_SECURITY_LABEL \ FATTR4_WORD2_SECURITY_LABEL
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 94/231 lockd: set fl_owner when unlocking files
In reply to Greg Kroah-HartmanFrom: Jeff Layton <[email protected]>
[ Upstream commit aec158242b87a43d83322e99bc71ab4428e5ab79 ]
Unlocking a POSIX lock on an inode with vfs_lock_file only works if the owner matches. Ensure we set it in the request.
--- fs/lockd/svcsubs.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
fs/lockd/svcsubs.c +3 -2
@@ -176,7 +176,7 @@ nlm_delete_file(struct nlm_file *file) } } -static int nlm_unlock_files (struct nlm_file *file) +static int nlm_unlock_files (struct nlm_file *file,fl_owner_t owner) {struct file_lock lock;
@@ -184,6 +184,7 @@ static int nlm_unlock_files(struct nlm_file *file) lock.fl_type =F_UNLCK ; lock.fl_start =0 ; lock.fl_end =OFFSET_MAX ; + lock.fl_owner = owner;if (file->f_file[O_RDONLY ] &&vfs_lock_file (file->f_file[O_RDONLY ],F_SETLK , &lock,NULL )) goto out_err;
@@ -225,7 +226,7 @@ nlm_traverse_locks(struct nlm_host *host, struct nlm_file *file,if (match (lockhost, host)) {spin_unlock (&flctx->flc_lock); -if (nlm_unlock_files (file)) +if (nlm_unlock_files (file, fl->fl_owner))return 1 ; goto again; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 95/231 lockd: fix nlm_close_files
In reply to Greg Kroah-HartmanFrom: Jeff Layton <[email protected]>
[ Upstream commit 1197eb5906a5464dbaea24cac296dfc38499cc00 ]
This loop condition tries a bit too hard to be clever. Just test for the two indices we care about explicitly.
--- fs/lockd/svcsubs.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-)
fs/lockd/svcsubs.c +4 -5
@@ -283,11 +283,10 @@ nlm_file_inuse(struct nlm_file *file)static void nlm_close_files (struct nlm_file *file) { -struct file *f; - -for (f = file->f_file[0 ]; f <= file->f_file[1 ]; f++) -if (f) - nlmsvc_ops->fclose (f); +if (file->f_file[O_RDONLY ]) + nlmsvc_ops->fclose (file->f_file[O_RDONLY ]); +if (file->f_file[O_WRONLY ]) + nlmsvc_ops->fclose (file->f_file[O_WRONLY ]); }/*
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 96/231 net: marvell: prestera: fix missed deinit sequence
In reply to Greg Kroah-HartmanFrom: Yevhen Orlov <[email protected]>
[ Upstream commit f946964a9f79f8dcb5a6329265281eebfc23aee5 ]
Add unregister_fib_notifier as rollback of register_fib_notifier.
--- drivers/net/ethernet/marvell/prestera/prestera_router.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/ethernet/marvell/prestera/prestera_router.c +1 -0
@@ -587,6 +587,7 @@ int prestera_router_init(struct prestera_switch *sw)void prestera_router_fini (struct prestera_switch *sw) { +unregister_fib_notifier (&init_net, &sw->router->fib_nb);unregister_inetaddr_notifier (&sw->router->inetaddr_nb);unregister_inetaddr_validator_notifier (&sw->router->inetaddr_valid_nb);rhashtable_destroy (&sw->router->kern_fib_cache_ht);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 97/231 ice: handle E822 generic device ID in PLDM header
In reply to Greg Kroah-HartmanFrom: Paul M Stillwell Jr <[email protected]>
[ Upstream commit f52d166819a4d8e0d5cca07d8a8dd6397c96dcf1 ]
The driver currently presumes that the record data in the PLDM header of the firmware image will match the device ID of the running device. This is true for E810 devices. It appears that for E822 devices that this is not guaranteed to be true.
Fix this by adding a check for the generic E822 device.
--- drivers/net/ethernet/intel/ice/ice_devids.h | 1 + .../net/ethernet/intel/ice/ice_fw_update.c | 96 ++++++++++++++++++- drivers/net/ethernet/intel/ice/ice_main.c | 1 + 3 files changed, 96 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/ice/ice_devids.h +1 -0
@@ -5,6 +5,7 @@#define _ICE_DEVIDS_H_/* Device IDs */ +#define ICE_DEV_ID_E822_SI_DFLT 0x1888/* Intel(R) Ethernet Connection E823-L for backplane */ #define ICE_DEV_ID_E823L_BACKPLANE 0x124C/* Intel(R) Ethernet Connection E823-L for SFP */
drivers/net/ethernet/intel/ice/ice_fw_update.c +94 -2
114 lines not on this page · raw message
drivers/net/ethernet/intel/ice/ice_main.c +1 -0
@@ -5397,6 +5397,7 @@ static const struct pci_device_id ice_pci_tbl[] = { { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_10G_BASE_T), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_1GBE), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_QSFP), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822_SI_DFLT), 0 },/* required last entry */ { 0, } };
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 98/231 ice: change devlink code to read NVM in blocks
In reply to Greg Kroah-HartmanFrom: Paul M Stillwell Jr <[email protected]>
[ Upstream commit 7b6f9462a3234c35cf808453d39a074a04e71de1 ]
When creating a snapshot of the NVM the driver needs to read the entire contents from the NVM and store it. The NVM reads are protected by a lock that is shared between the driver and the firmware.
If the driver takes too long to read the entire NVM (which can happen on some systems) then the firmware could reclaim the lock and cause subsequent reads from the driver to fail.
We could fix this by increasing the timeout that we pass to the firmware, but we could end up in the same situation again if the system is slow. Instead have the driver break the reading of the NVM into blocks that are small enough that we have confidence that the read will complete within the timeout time, but large enough not to cause significant AQ overhead.
--- drivers/net/ethernet/intel/ice/ice_devlink.c | 59 +++++++++++++------- 1 file changed, 40 insertions(+), 19 deletions(-)
drivers/net/ethernet/intel/ice/ice_devlink.c +40 -19
@@ -792,6 +792,8 @@ void ice_devlink_destroy_vf_port(struct ice_vf *vf)devlink_port_unregister (devlink_port); } +#define ICE_DEVLINK_READ_BLK_SIZE (1024 * 1024) +/** * ice_devlink_nvm_snapshot - Capture a snapshot of the NVM flash contents * @devlink: the devlink instance
@@ -818,8 +820,9 @@ static int ice_devlink_nvm_snapshot(struct devlink *devlink,struct ice_pf *pf =devlink_priv (devlink);struct device *dev =ice_pf_to_dev (pf);struct ice_hw *hw = &pf->hw; -void *nvm_data; -u32 nvm_size; +u8 *nvm_data, *tmp, i; +u32 nvm_size, left; +s8 num_blks;int status; nvm_size = hw->flash.flash_size;
@@ -827,26 +830,44 @@ static int ice_devlink_nvm_snapshot(struct devlink *devlink,if (!nvm_data)return -ENOMEM ; - status =ice_acquire_nvm (hw,ICE_RES_READ ); -if (status) { -dev_dbg (dev,"ice_acquire_nvm failed, err %d aq_err %d\n" , - status, hw->adminq.sq_last_status); -NL_SET_ERR_MSG_MOD (extack,"Failed to acquire NVM semaphore" ); -vfree (nvm_data); -return status; - } - status =ice_read_flat_nvm (hw,0 , &nvm_size, nvm_data, false); -if (status) { -dev_dbg (dev,"ice_read_flat_nvm failed after reading %u bytes, err %d aq_err %d\n" , - nvm_size, status, hw->adminq.sq_last_status); -NL_SET_ERR_MSG_MOD (extack,"Failed to read NVM contents" ); + num_blks =DIV_ROUND_UP (nvm_size,ICE_DEVLINK_READ_BLK_SIZE ); + tmp = nvm_data; + left = nvm_size; + +/* Some systems take longer to read the NVM than others which causes the +* FW to reclaim the NVM lock before the entire NVM has been read. Fix +* this by breaking the reads of the NVM into smaller chunks that will +* probably not take as long. This has some overhead since we are +* increasing the number of AQ commands, but it should always work +*/ +for (i =0 ; i < num_blks; i++) { +u32 read_sz =min_t (u32,ICE_DEVLINK_READ_BLK_SIZE , left); + + status =ice_acquire_nvm (hw,ICE_RES_READ ); +if (status) { +dev_dbg (dev,"ice_acquire_nvm failed, err %d aq_err %d\n" , + status, hw->adminq.sq_last_status); +NL_SET_ERR_MSG_MOD (extack,"Failed to acquire NVM semaphore" ); +vfree (nvm_data); +return -EIO ; + } + + status =ice_read_flat_nvm (hw, i *ICE_DEVLINK_READ_BLK_SIZE , + &read_sz, tmp, false); +if (status) { +dev_dbg (dev,"ice_read_flat_nvm failed after reading %u bytes, err %d aq_err %d\n" , + read_sz, status, hw->adminq.sq_last_status); +NL_SET_ERR_MSG_MOD (extack,"Failed to read NVM contents" ); +ice_release_nvm (hw); +vfree (nvm_data); +return -EIO ; + }ice_release_nvm (hw); -vfree (nvm_data); -return status; - } -ice_release_nvm (hw); + tmp += read_sz; + left -= read_sz; + } *data = nvm_data;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 99/231 tracing: Fix sleeping while atomic in kdb ftdump
In reply to Greg Kroah-HartmanFrom: Douglas Anderson <[email protected]>
[ Upstream commit 495fcec8648cdfb483b5b9ab310f3839f07cb3b8 ]
If you drop into kdb and type "ftdump" you'll get a sleeping while atomic warning from memory allocation in trace_find_next_entry().
This appears to have been caused by commit ff895103a84a ("tracing:
Save off entry when peeking at next entry"), which added the allocation in that path. The problematic commit was already fixed by commit 8e99cf91b99b ("tracing: Do not allocate buffer in trace_find_next_entry() in atomic") but that fix missed the kdb case.
The fix here is easy: just move the assignment of the static buffer to the place where it should have been to begin with:trace_init_global_iter(). That function is called in two places, once is right before the assignment of the static buffer added by the previous fix and once is in kdb.
Note that it appears that there's a second static buffer that we need to assign that was added in commit efbbdaa22bb7 ("tracing: Show real address for trace event arguments"), so we'll move that too.
--- kernel/trace/trace.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
kernel/trace/trace.c +6 -5
@@ -9863,6 +9863,12 @@ void trace_init_global_iter(struct trace_iterator *iter)/* Output in nanoseconds only if we are using a clock in nanoseconds. */ if (trace_clocks[iter->tr->clock_id].in_ns) iter->iter_flags |=TRACE_FILE_TIME_IN_NS ; + +/* Can not use kmalloc for iter.temp and iter.fmt */ + iter->temp = static_temp_buf; + iter->temp_size =STATIC_TEMP_BUF_SIZE ; + iter->fmt = static_fmt_buf; + iter->fmt_size =STATIC_FMT_BUF_SIZE ; }void ftrace_dump (enum ftrace_dump_mode oops_dump_mode)
@@ -9895,11 +9901,6 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode)/* Simulate the iterator */ trace_init_global_iter (&iter); -/* Can not use kmalloc for iter.temp and iter.fmt */ - iter.temp = static_temp_buf; - iter.temp_size =STATIC_TEMP_BUF_SIZE ; - iter.fmt = static_fmt_buf; - iter.fmt_size =STATIC_FMT_BUF_SIZE ;for_each_tracing_cpu (cpu) {atomic_inc (&per_cpu_ptr (iter.array_buffer->data, cpu)->disabled);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 100/231 drm/i915/selftests: fix a couple IS_ERR() vs NULL tests
In reply to Greg Kroah-HartmanFrom: Dan Carpenter <[email protected]>
[ Upstream commit 896dcabd1f8f613c533d948df17408c41f8929f5 ]
The shmem_pin_map() function doesn't return error pointers, it returns NULL.
--- drivers/gpu/drm/i915/gt/selftest_lrc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/gpu/drm/i915/gt/selftest_lrc.c +4 -4
@@ -155,8 +155,8 @@ static int live_lrc_layout(void *arg)continue ; hw =shmem_pin_map (engine->default_state); -if (IS_ERR (hw)) { - err =PTR_ERR (hw); +if (!hw) { + err = -ENOMEM ;break ; } hw +=LRC_STATE_OFFSET /sizeof (*hw);
@@ -331,8 +331,8 @@ static int live_lrc_fixed(void *arg)continue ; hw =shmem_pin_map (engine->default_state); -if (IS_ERR (hw)) { - err =PTR_ERR (hw); +if (!hw) { + err = -ENOMEM ;break ; } hw +=LRC_STATE_OFFSET /sizeof (*hw);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 101/231 drm/i915/ttm: fix sg_table construction
In reply to Greg Kroah-HartmanFrom: Matthew Auld <[email protected]>
[ Upstream commit aff1e0b09b54b64944b7fe32997229552737b9e9 ]
If we encounter some monster sized local-memory page that exceeds the maximum sg length (UINT32_MAX), ensure that don't end up with some misaligned address in the entry that follows, leading to fireworks later. Also ensure we have some coverage of this in the selftests.
v2(Chris):
- Use round_down consistently to avoid udiv errors
v3(Nirmoy):
- Also update the max_segment in the selftest
--- drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 11 ++++++++-- drivers/gpu/drm/i915/i915_scatterlist.c | 19 +++++++++++++---- drivers/gpu/drm/i915/i915_scatterlist.h | 6 ++++-- drivers/gpu/drm/i915/intel_region_ttm.c | 10 ++++++--- drivers/gpu/drm/i915/intel_region_ttm.h | 3 ++- .../drm/i915/selftests/intel_memory_region.c | 21 +++++++++++++++++-- drivers/gpu/drm/i915/selftests/mock_region.c | 3 ++- 7 files changed, 58 insertions(+), 15 deletions(-)
drivers/gpu/drm/i915/gem/i915_gem_ttm.c +9 -2
@@ -583,10 +583,15 @@ i915_ttm_resource_get_st(struct drm_i915_gem_object *obj, struct ttm_resource *res) {struct ttm_buffer_object *bo =i915_gem_to_ttm (obj); +u64 page_alignment;if (!i915_ttm_gtt_binds_lmem (res))return i915_ttm_tt_get_st (bo->ttm); + page_alignment = bo->page_alignment <<PAGE_SHIFT ; +if (!page_alignment) + page_alignment = obj->mm.region->min_page_size; +/* * If CPU mapping differs, we need to add the ttm_tt pages to * the resulting st. Might make sense for GGTT.
@@ -597,7 +602,8 @@ i915_ttm_resource_get_st(struct drm_i915_gem_object *obj,struct i915_refct_sgt *rsgt; rsgt =intel_region_ttm_resource_to_rsgt (obj->mm.region, - res); + res, + page_alignment);if (IS_ERR (rsgt))return rsgt;
@@ -606,7 +612,8 @@ i915_ttm_resource_get_st(struct drm_i915_gem_object *obj,return i915_refct_sgt_get (obj->ttm.cached_io_rsgt); } -return intel_region_ttm_resource_to_rsgt (obj->mm.region, res); +return intel_region_ttm_resource_to_rsgt (obj->mm.region, res, + page_alignment); }static int i915_ttm_truncate (struct drm_i915_gem_object *obj)
drivers/gpu/drm/i915/i915_scatterlist.c +15 -4
@@ -68,6 +68,7 @@ void i915_refct_sgt_init(struct i915_refct_sgt *rsgt, size_t size)* drm_mm_node * @node: The drm_mm_node. * @region_start: An offset to add to the dma addresses of the sg list. +* @page_alignment: Required page alignment for each sg entry. Power of two. * * Create a struct sg_table, initializing it from a struct drm_mm_node, * taking a maximum segment length into account, splitting into segments
@@ -77,15 +78,18 @@ void i915_refct_sgt_init(struct i915_refct_sgt *rsgt, size_t size)* error code cast to an error pointer on failure. */ struct i915_refct_sgt *i915_rsgt_from_mm_node (const struct drm_mm_node *node, -u64 region_start) +u64 region_start, +u64 page_alignment) { -const u64 max_segment =SZ_1G ;/* Do we have a limit on this? */ +const u64 max_segment =round_down (UINT_MAX , page_alignment);u64 segment_pages = max_segment >>PAGE_SHIFT ;u64 block_size, offset, prev_end;struct i915_refct_sgt *rsgt;struct sg_table *st;struct scatterlist *sg; +GEM_BUG_ON (!max_segment); + rsgt =kmalloc (sizeof (*rsgt),GFP_KERNEL );if (!rsgt)return ERR_PTR (-ENOMEM );
@@ -112,6 +116,8 @@ struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node, sg =__sg_next (sg);sg_dma_address (sg) = region_start + offset; +GEM_BUG_ON (!IS_ALIGNED (sg_dma_address (sg), + page_alignment));sg_dma_len (sg) =0 ; sg->length =0 ; st->nents++;
@@ -138,6 +144,7 @@ struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node,* i915_buddy_block list * @res: The struct i915_ttm_buddy_resource. * @region_start: An offset to add to the dma addresses of the sg list. +* @page_alignment: Required page alignment for each sg entry. Power of two. * * Create a struct sg_table, initializing it from struct i915_buddy_block list, * taking a maximum segment length into account, splitting into segments
@@ -147,11 +154,12 @@ struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node,* error code cast to an error pointer on failure. */ struct i915_refct_sgt *i915_rsgt_from_buddy_resource (struct ttm_resource *res, -u64 region_start) +u64 region_start, +u64 page_alignment) {struct i915_ttm_buddy_resource *bman_res =to_ttm_buddy_resource (res);const u64 size = res->num_pages <<PAGE_SHIFT ; -const u64 max_segment =rounddown (UINT_MAX ,PAGE_SIZE ); +const u64 max_segment =round_down (UINT_MAX , page_alignment);struct drm_buddy *mm = bman_res->mm;struct list_head *blocks = &bman_res->blocks;struct drm_buddy_block *block;
@@ -161,6 +169,7 @@ struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res,resource_size_t prev_end;GEM_BUG_ON (list_empty (blocks)); +GEM_BUG_ON (!max_segment); rsgt =kmalloc (sizeof (*rsgt),GFP_KERNEL );if (!rsgt)
@@ -191,6 +200,8 @@ struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res, sg =__sg_next (sg);sg_dma_address (sg) = region_start + offset; +GEM_BUG_ON (!IS_ALIGNED (sg_dma_address (sg), + page_alignment));sg_dma_len (sg) =0 ; sg->length =0 ; st->nents++;
drivers/gpu/drm/i915/i915_scatterlist.h +4 -2
@@ -213,9 +213,11 @@ static inline void __i915_refct_sgt_init(struct i915_refct_sgt *rsgt,void i915_refct_sgt_init (struct i915_refct_sgt *rsgt,size_t size);struct i915_refct_sgt *i915_rsgt_from_mm_node (const struct drm_mm_node *node, -u64 region_start); +u64 region_start, +u64 page_alignment);struct i915_refct_sgt *i915_rsgt_from_buddy_resource (struct ttm_resource *res, -u64 region_start); +u64 region_start, +u64 page_alignment);#endif
drivers/gpu/drm/i915/intel_region_ttm.c +7 -3
@@ -151,6 +151,7 @@ int intel_region_ttm_fini(struct intel_memory_region *mem)* Convert an opaque TTM resource manager resource to a refcounted sg_table. * @mem: The memory region. * @res: The resource manager resource obtained from the TTM resource manager. +* @page_alignment: Required page alignment for each sg entry. Power of two. * * The gem backends typically use sg-tables for operations on the underlying * io_memory. So provide a way for the backends to translate the
@@ -160,16 +161,19 @@ int intel_region_ttm_fini(struct intel_memory_region *mem)*/ struct i915_refct_sgt *intel_region_ttm_resource_to_rsgt (struct intel_memory_region *mem, -struct ttm_resource *res) +struct ttm_resource *res, +u64 page_alignment) {if (mem->is_range_manager) {struct ttm_range_mgr_node *range_node =to_ttm_range_mgr_node (res);return i915_rsgt_from_mm_node (&range_node->mm_nodes[0 ], - mem->region.start); + mem->region.start, + page_alignment); }else { -return i915_rsgt_from_buddy_resource (res, mem->region.start); +return i915_rsgt_from_buddy_resource (res, mem->region.start, + page_alignment); } }
drivers/gpu/drm/i915/intel_region_ttm.h +2 -1
@@ -24,7 +24,8 @@ int intel_region_ttm_fini(struct intel_memory_region *mem);struct i915_refct_sgt *intel_region_ttm_resource_to_rsgt (struct intel_memory_region *mem, -struct ttm_resource *res); +struct ttm_resource *res, +u64 page_alignment);void intel_region_ttm_resource_free (struct intel_memory_region *mem,struct ttm_resource *res);
drivers/gpu/drm/i915/selftests/intel_memory_region.c +19 -2
@@ -451,7 +451,6 @@ static int igt_mock_splintered_region(void *arg)static int igt_mock_max_segment (void *arg) { -const unsigned int max_segment =rounddown (UINT_MAX ,PAGE_SIZE );struct intel_memory_region *mem = arg;struct drm_i915_private *i915 = mem->i915;struct i915_ttm_buddy_resource *res;
@@ -460,7 +459,10 @@ static int igt_mock_max_segment(void *arg)struct drm_buddy *mm;struct list_head *blocks;struct scatterlist *sg; +I915_RND_STATE (prng);LIST_HEAD (objects); +unsigned int max_segment; +unsigned int ps;u64 size;int err =0 ;
@@ -472,7 +474,13 @@ static int igt_mock_max_segment(void *arg)*/ size =SZ_8G ; - mem =mock_region_create (i915,0 , size,PAGE_SIZE ,0 ,0 ); + ps =PAGE_SIZE ; +if (i915_prandom_u64_state (&prng) &1 ) + ps =SZ_64K ;/* For something like DG2 */ + + max_segment =round_down (UINT_MAX , ps); + + mem =mock_region_create (i915,0 , size, ps,0 ,0 );if (IS_ERR (mem))return PTR_ERR (mem);
@@ -498,12 +506,21 @@ static int igt_mock_max_segment(void *arg) }for (sg = obj->mm.pages->sgl; sg; sg =sg_next (sg)) { +dma_addr_t daddr =sg_dma_address (sg); +if (sg->length > max_segment) {pr_err ("%s: Created an oversized scatterlist entry, %u > %u\n" , __func__, sg->length, max_segment); err = -EINVAL ; goto out_close; } + +if (!IS_ALIGNED (daddr, ps)) { +pr_err ("%s: Created an unaligned scatterlist entry, addr=%pa, ps=%u\n" , + __func__, &daddr, ps); + err = -EINVAL ; + goto out_close; + } } out_close:
drivers/gpu/drm/i915/selftests/mock_region.c +2 -1
@@ -32,7 +32,8 @@ static int mock_region_get_pages(struct drm_i915_gem_object *obj)return PTR_ERR (obj->mm.res); obj->mm.rsgt =intel_region_ttm_resource_to_rsgt (obj->mm.region, - obj->mm.res); + obj->mm.res, + obj->mm.region->min_page_size);if (IS_ERR (obj->mm.rsgt)) { err =PTR_ERR (obj->mm.rsgt); goto err_free_resource;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 102/231 drm/i915/gt: Serialize GRDOM access between multiple engine resets
In reply to Greg Kroah-HartmanFrom: Chris Wilson <[email protected]>
[ Upstream commit b24dcf1dc507f69ed3b5c66c2b6a0209ae80d4d4 ]
Don't allow two engines to be reset in parallel, as they would both try to select a reset bit (and send requests to common registers) and wait on that register, at the same time. Serialize control of the reset requests/acks using the uncore->lock, which will also ensure that no other GT state changes at the same time as the actual reset.
--- drivers/gpu/drm/i915/gt/intel_reset.c | 37 ++++++++++++++++++++------- 1 file changed, 28 insertions(+), 9 deletions(-)
drivers/gpu/drm/i915/gt/intel_reset.c +28 -9
@@ -299,9 +299,9 @@ static int gen6_hw_domain_reset(struct intel_gt *gt, u32 hw_domain_mask)return err; } -static int gen6_reset_engines (struct intel_gt *gt, -intel_engine_mask_t engine_mask, -unsigned int retry) +static int __gen6_reset_engines (struct intel_gt *gt, +intel_engine_mask_t engine_mask, +unsigned int retry) {struct intel_engine_cs *engine;u32 hw_mask;
@@ -320,6 +320,20 @@ static int gen6_reset_engines(struct intel_gt *gt,return gen6_hw_domain_reset (gt, hw_mask); } +static int gen6_reset_engines (struct intel_gt *gt, +intel_engine_mask_t engine_mask, +unsigned int retry) +{ +unsigned long flags; +int ret; + +spin_lock_irqsave (>->uncore->lock, flags); + ret =__gen6_reset_engines (gt, engine_mask, retry); +spin_unlock_irqrestore (>->uncore->lock, flags); + +return ret; +} +static struct intel_engine_cs *find_sfc_paired_vecs_engine (struct intel_engine_cs *engine) {int vecs_id;
@@ -486,9 +500,9 @@ static void gen11_unlock_sfc(struct intel_engine_cs *engine)rmw_clear_fw (uncore, sfc_lock.lock_reg, sfc_lock.lock_bit); } -static int gen11_reset_engines (struct intel_gt *gt, -intel_engine_mask_t engine_mask, -unsigned int retry) +static int __gen11_reset_engines (struct intel_gt *gt, +intel_engine_mask_t engine_mask, +unsigned int retry) {struct intel_engine_cs *engine;intel_engine_mask_t tmp;
@@ -582,8 +596,11 @@ static int gen8_reset_engines(struct intel_gt *gt,struct intel_engine_cs *engine;const bool reset_non_ready = retry >=1 ;intel_engine_mask_t tmp; +unsigned long flags;int ret; +spin_lock_irqsave (>->uncore->lock, flags); +for_each_engine_masked (engine, gt, engine_mask, tmp) { ret =gen8_engine_reset_prepare (engine);if (ret && !reset_non_ready)
@@ -611,17 +628,19 @@ static int gen8_reset_engines(struct intel_gt *gt,* This is best effort, so ignore any error from the initial reset. */ if (IS_DG2 (gt->i915) && engine_mask ==ALL_ENGINES ) -gen11_reset_engines (gt, gt->info.engine_mask,0 ); +__gen11_reset_engines (gt, gt->info.engine_mask,0 );if (GRAPHICS_VER (gt->i915) >=11 ) - ret =gen11_reset_engines (gt, engine_mask, retry); + ret =__gen11_reset_engines (gt, engine_mask, retry);else - ret =gen6_reset_engines (gt, engine_mask, retry); + ret =__gen6_reset_engines (gt, engine_mask, retry); skip_reset:for_each_engine_masked (engine, gt, engine_mask, tmp)gen8_engine_reset_cancel (engine); +spin_unlock_irqrestore (>->uncore->lock, flags); +return ret; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 103/231 drm/i915/gt: Serialize TLB invalidates with GT resets
In reply to Greg Kroah-HartmanFrom: Chris Wilson <[email protected]>
[ Upstream commit a1c5a7bf79c1faa5633b918b5c0666545e84c4d1 ]
Avoid trying to invalidate the TLB in the middle of performing an engine reset, as this may result in the reset timing out. Currently, the TLB invalidate is only serialised by its own mutex, forgoing the uncore lock, but we can take the uncore->lock as well to serialise the mmio access, thereby serialising with the GDRST.
Tested on a NUC5i7RYB, BIOS RYBDWi35.86A.0380.2019.0517.1530 with i915 selftest/hangcheck.
--- drivers/gpu/drm/i915/gt/intel_gt.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/gt/intel_gt.c +14 -1
@@ -1012,6 +1012,20 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt)mutex_lock (>->tlb_invalidate_lock);intel_uncore_forcewake_get (uncore,FORCEWAKE_ALL ); +spin_lock_irq (&uncore->lock);/* serialise invalidate with GT reset */ + +for_each_engine (engine, gt, id) { +struct reg_and_bit rb; + + rb =get_reg_and_bit (engine, regs == gen8_regs, regs, num); +if (!i915_mmio_reg_offset (rb.reg)) +continue ; + +intel_uncore_write_fw (uncore, rb.reg, rb.bit); + } + +spin_unlock_irq (&uncore->lock); +for_each_engine (engine, gt, id) {/* * HW architecture suggest typical invalidation time at 40us,
@@ -1026,7 +1040,6 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt)if (!i915_mmio_reg_offset (rb.reg))continue ; -intel_uncore_write_fw (uncore, rb.reg, rb.bit); if (__intel_wait_for_register_fw(uncore, rb.reg, rb.bit, 0, timeout_us, timeout_ms,
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 104/231 drm/i915/selftests: fix subtraction overflow bug
In reply to Greg Kroah-HartmanFrom: Andrzej Hajda <[email protected]>
[ Upstream commit 333991c4e66b3d4b5613315f18016da80344f659 ]
On some machines hole_end can be small enough to cause subtraction overflow. On the other side (addr + 2 * min_alignment) can overflow in case of mock tests. This patch should handle both cases.
--- drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +1 -1
@@ -742,7 +742,7 @@ static int pot_hole(struct i915_address_space *vm,u64 addr;for (addr =round_up (hole_start + min_alignment, step) - min_alignment; - addr <=round_down (hole_end - (2 * min_alignment), step) - min_alignment; + hole_end > addr && hole_end - addr >=2 * min_alignment; addr += step) { err =i915_vma_pin (vma,0 ,0 , addr | flags);if (err) {
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 105/231 bnxt_en: reclaim max resources if sriov enable fails
In reply to Greg Kroah-HartmanFrom: Kashyap Desai <[email protected]>
[ Upstream commit c5b744d38c36a407a41e918602eec4d89730787b ]
If bnxt_sriov_enable() fails after some resources have been reserved for the VFs, the current code is not unwinding properly and the reserved resources become unavailable afterwards. Fix it by properly unwinding with a call to bnxt_hwrm_func_qcaps() to reset all maximum resources.
Also, add the missing bnxt_ulp_sriov_cfg() call to let the RDMA driver know to abort.
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 + drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 7 ++++++- 3 files changed, 8 insertions(+), 2 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +1 -1
@@ -7641,7 +7641,7 @@ static void bnxt_hwrm_dbg_qcaps(struct bnxt *bp)static int bnxt_hwrm_queue_qportcfg (struct bnxt *bp); -static int bnxt_hwrm_func_qcaps (struct bnxt *bp) +int bnxt_hwrm_func_qcaps (struct bnxt *bp) {int rc;
drivers/net/ethernet/broadcom/bnxt/bnxt.h +1 -0
@@ -2310,6 +2310,7 @@ int bnxt_cancel_reservations(struct bnxt *bp, bool fw_reset);int bnxt_hwrm_alloc_wol_fltr (struct bnxt *bp);int bnxt_hwrm_free_wol_fltr (struct bnxt *bp);int bnxt_hwrm_func_resc_qcaps (struct bnxt *bp,bool all); +int bnxt_hwrm_func_qcaps (struct bnxt *bp);int bnxt_hwrm_fw_set_time (struct bnxt *);int bnxt_open_nic (struct bnxt *,bool ,bool );int bnxt_half_open_nic (struct bnxt *bp);
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +6 -1
@@ -823,8 +823,10 @@ static int bnxt_sriov_enable(struct bnxt *bp, int *num_vfs) goto err_out2; rc =pci_enable_sriov (bp->pdev, *num_vfs); -if (rc) +if (rc) { +bnxt_ulp_sriov_cfg (bp,0 ); goto err_out2; + }return 0 ;
@@ -832,6 +834,9 @@ static int bnxt_sriov_enable(struct bnxt *bp, int *num_vfs)/* Free the resources reserved for various VF's */ bnxt_hwrm_func_vf_resource_free (bp, *num_vfs); +/* Restore the max resources */ +bnxt_hwrm_func_qcaps (bp); + err_out1:bnxt_free_vf_resources (bp);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 106/231 bnxt_en: Fix bnxt_reinit_after_abort() code path
In reply to Greg Kroah-HartmanFrom: Michael Chan <[email protected]>
[ Upstream commit 4279414bff8af9898e8c53ae6c5bc17f68ad67b7 ]
bnxt_reinit_after_abort() is called during ifup when a previous FW reset sequence has aborted or a previous ifup has failed after detecting FW reset. In all cases, it is safe to assume that a previous FW reset has completed and the driver may not have fully reinitialized.
Prior to this patch, it is assumed that the FUNC_DRV_IF_CHANGE_RESP_FLAGS_HOT_FW_RESET_DONE flag will always be set by the firmware in bnxt_hwrm_if_change(). This may not be true if the driver has already attempted to register with the firmware. The firmware may not set the RESET_DONE flag again after the driver has registered, assuming that the driver has seen the flag already.
Fix it to always go through the FW reset initialization path if the BNXT_STATE_FW_RESET_DET flag is set. This flag is always set by the driver after successfully going through bnxt_reinit_after_abort().
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +2 -1
@@ -9916,7 +9916,8 @@ static int bnxt_hwrm_if_change(struct bnxt *bp, bool up)if (flags &FUNC_DRV_IF_CHANGE_RESP_FLAGS_RESC_CHANGE ) resc_reinit = true; -if (flags &FUNC_DRV_IF_CHANGE_RESP_FLAGS_HOT_FW_RESET_DONE ) +if (flags &FUNC_DRV_IF_CHANGE_RESP_FLAGS_HOT_FW_RESET_DONE || +test_bit (BNXT_STATE_FW_RESET_DET , &bp->state)) fw_reset = true;else bnxt_remap_fw_health_regs (bp);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 107/231 bnxt_en: fix livepatch query
In reply to Greg Kroah-HartmanFrom: Vikas Gupta <[email protected]>
[ Upstream commit 619b9b1622c283cc5ca86f4c487db266a8f55dab ]
In the livepatch query fw_target BNXT_FW_SRT_PATCH is applicable for P5 chips only.
--- drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c +5 -3
@@ -979,9 +979,11 @@ static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req,if (rc)return rc; - rc =bnxt_dl_livepatch_info_put (bp, req,BNXT_FW_SRT_PATCH ); -if (rc) -return rc; +if (BNXT_CHIP_P5 (bp)) { + rc =bnxt_dl_livepatch_info_put (bp, req,BNXT_FW_SRT_PATCH ); +if (rc) +return rc; + }return bnxt_dl_livepatch_info_put (bp, req,BNXT_FW_CRT_PATCH ); }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 108/231 bnxt_en: Fix bnxt_refclk_read()
In reply to Greg Kroah-HartmanFrom: Pavan Chebbi <[email protected]>
[ Upstream commit ddde5412fdaa5048bbca31529d46cb8da882870c ]
The upper 32-bit PHC register is not latched when reading the lower 32-bit PHC register. Current code leaves a small window where we may not read correct higher order bits if the lower order bits are just about to wrap around.
This patch fixes this by reading higher order bits twice and makes sure that final value is correctly paired with its lower 32 bits.
--- drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c +11 -2
@@ -76,14 +76,23 @@ static int bnxt_refclk_read(struct bnxt *bp, struct ptp_system_timestamp *sts, u64 *ns) {struct bnxt_ptp_cfg *ptp = bp->ptp_cfg; +u32 high_before, high_now, low;if (test_bit (BNXT_STATE_IN_FW_RESET , &bp->state))return -EIO ; + high_before =readl (bp->bar0 + ptp->refclk_mapped_regs[1 ]);ptp_read_system_prets (sts); - *ns =readl (bp->bar0 + ptp->refclk_mapped_regs[0 ]); + low =readl (bp->bar0 + ptp->refclk_mapped_regs[0 ]);ptp_read_system_postts (sts); - *ns |= (u64 )readl (bp->bar0 + ptp->refclk_mapped_regs[1 ]) <<32 ; + high_now =readl (bp->bar0 + ptp->refclk_mapped_regs[1 ]); +if (high_now != high_before) { +ptp_read_system_prets (sts); + low =readl (bp->bar0 + ptp->refclk_mapped_regs[0 ]); +ptp_read_system_postts (sts); + } + *ns = ((u64 )high_now <<32 ) | low; +return 0 ; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 109/231 sysctl: Fix data-races in proc_dou8vec_minmax().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 7dee5d7747a69aa2be41f04c6a7ecfe3ac8cdf18 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_dou8vec_minmax() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_dou8vec_minmax() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
kernel/sysctl.c +2 -2
@@ -1079,13 +1079,13 @@ int proc_dou8vec_minmax(struct ctl_table *table, int write, tmp.maxlen =sizeof (val); tmp.data = &val; - val = *data; + val =READ_ONCE (*data); res =do_proc_douintvec (&tmp, write, buffer, lenp, ppos, do_proc_douintvec_minmax_conv, ¶m);if (res)return res;if (write) - *data = val; +WRITE_ONCE (*data, val);return 0 ; }EXPORT_SYMBOL_GPL (proc_dou8vec_minmax);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 110/231 sysctl: Fix data-races in proc_dointvec_ms_jiffies().
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 7d1025e559782b58824b36cb8ad547a69f2e4b31 ]
A sysctl variable is accessed concurrently, and there is always a chance of data-race. So, all readers and writers need some basic protection to avoid load/store-tearing.
This patch changes proc_dointvec_ms_jiffies() to use READ_ONCE() and WRITE_ONCE() internally to fix data-races on the sysctl side. For now, proc_dointvec_ms_jiffies() itself is tolerant to a data-race, but we still need to add annotations on the other subsystem's side.
--- kernel/sysctl.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
kernel/sysctl.c +4 -4
@@ -1296,9 +1296,9 @@ static int do_proc_dointvec_ms_jiffies_conv(bool *negp, unsigned long *lvalp,if (jif >INT_MAX )return 1 ; - *valp = (int )jif; +WRITE_ONCE (*valp, (int )jif); }else { -int val = *valp; +int val =READ_ONCE (*valp);unsigned long lval;if (val <0 ) { *negp = true;
@@ -1366,8 +1366,8 @@ int proc_dointvec_userhz_jiffies(struct ctl_table *table, int write,* @ppos: the current position in the file * * Reads/writes up to table->maxlen/sizeof(unsigned int) integer -* values from/to the user buffer, treated as an ASCII string. -* The values read are assumed to be in 1/1000 seconds, and +* values from/to the user buffer, treated as an ASCII string. +* The values read are assumed to be in 1/1000 seconds, and * are converted into jiffies. * * Returns 0 on success.
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 111/231 tcp: Fix a data-race around sysctl_max_tw_buckets.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 6f605b57f3782114e330e108ce1903ede22ec675 ]
While reading sysctl_max_tw_buckets, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/inet_timewait_sock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/ipv4/inet_timewait_sock.c +2 -1
@@ -156,7 +156,8 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk, {struct inet_timewait_sock *tw; -if (refcount_read (&dr->tw_refcount) -1 >= dr->sysctl_max_tw_buckets) +if (refcount_read (&dr->tw_refcount) -1 >= +READ_ONCE (dr->sysctl_max_tw_buckets))return NULL ; tw = kmem_cache_alloc(sk->sk_prot_creator->twsk_prot->twsk_slab,
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 112/231 icmp: Fix a data-race around sysctl_icmp_echo_ignore_all.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit bb7bb35a63b4812da8e3aff587773678e31d23e3 ]
While reading sysctl_icmp_echo_ignore_all, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/icmp.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-)
net/ipv4/icmp.c +1 -1
@@ -990,7 +990,7 @@ static bool icmp_echo(struct sk_buff *skb) net =dev_net (skb_dst (skb)->dev);/* should there be an ICMP stat for ignored echos? */ -if (net->ipv4.sysctl_icmp_echo_ignore_all) +if (READ_ONCE (net->ipv4.sysctl_icmp_echo_ignore_all))return true; icmp_param.data.icmph = *icmp_hdr (skb);
net/ipv4/sysctl_net_ipv4.c +2 -0
@@ -603,6 +603,8 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 =SYSCTL_ONE }, { .procname = "icmp_echo_enable_probe",
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 113/231 icmp: Fix data-races around sysctl_icmp_echo_enable_probe.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 4a2f7083cc6cb72dade9a63699ca352fad26d1cd ]
While reading sysctl_icmp_echo_enable_probe, it can be changed concurrently. Thus, we need to add READ_ONCE() to its readers.
--- net/ipv4/icmp.c | 2 +- net/ipv6/icmp.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
net/ipv4/icmp.c +1 -1
@@ -1025,7 +1025,7 @@ bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr)u16 ident_len;u8 status; -if (!net->ipv4.sysctl_icmp_echo_enable_probe) +if (!READ_ONCE (net->ipv4.sysctl_icmp_echo_enable_probe))return false;/* We currently only support probing interfaces on the proxy node
net/ipv6/icmp.c +1 -1
@@ -919,7 +919,7 @@ static int icmpv6_rcv(struct sk_buff *skb)break ;case ICMPV6_EXT_ECHO_REQUEST :if (!net->ipv6.sysctl.icmpv6_echo_ignore_all && - net->ipv4.sysctl_icmp_echo_enable_probe) +READ_ONCE (net->ipv4.sysctl_icmp_echo_enable_probe))icmpv6_echo_reply (skb);break ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 114/231 icmp: Fix a data-race around sysctl_icmp_echo_ignore_broadcasts.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 66484bb98ed2dfa1dda37a32411483d8311ac269 ]
While reading sysctl_icmp_echo_ignore_broadcasts, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/icmp.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-)
net/ipv4/icmp.c +1 -1
@@ -1239,7 +1239,7 @@ int icmp_rcv(struct sk_buff *skb)*/ if ((icmph->type ==ICMP_ECHO || icmph->type ==ICMP_TIMESTAMP ) && - net->ipv4.sysctl_icmp_echo_ignore_broadcasts) { +READ_ONCE (net->ipv4.sysctl_icmp_echo_ignore_broadcasts)) { goto error; } if (icmph->type != ICMP_ECHO &&
net/ipv4/sysctl_net_ipv4.c +2 -0
@@ -621,6 +621,8 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 =SYSCTL_ONE }, { .procname = "icmp_ignore_bogus_error_responses",
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 115/231 icmp: Fix a data-race around sysctl_icmp_ignore_bogus_error_responses.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit b04f9b7e85c7d7aecbada620e8759a662af068d3 ]
While reading sysctl_icmp_ignore_bogus_error_responses, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/icmp.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-)
net/ipv4/icmp.c +1 -1
@@ -930,7 +930,7 @@ static bool icmp_unreach(struct sk_buff *skb)* get the other vendor to fix their kit. */ -if (!net->ipv4.sysctl_icmp_ignore_bogus_error_responses && +if (!READ_ONCE (net->ipv4.sysctl_icmp_ignore_bogus_error_responses) &&inet_addr_type_dev_table (net, skb->dev, iph->daddr) ==RTN_BROADCAST ) { net_warn_ratelimited("%pI4 sent an invalid ICMP type %u, code %u error to a broadcast: %pI4 on %s\n", &ip_hdr(skb)->saddr,
net/ipv4/sysctl_net_ipv4.c +2 -0
@@ -630,6 +630,8 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 =SYSCTL_ONE }, { .procname = "icmp_errors_use_inbound_ifaddr",
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 116/231 icmp: Fix a data-race around sysctl_icmp_errors_use_inbound_ifaddr.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit d2efabce81db7eed1c98fa1a3f203f0edd738ac3 ]
While reading sysctl_icmp_errors_use_inbound_ifaddr, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/icmp.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-)
net/ipv4/icmp.c +1 -1
@@ -693,7 +693,7 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info,rcu_read_lock ();if (rt_is_input_route (rt) && - net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr) +READ_ONCE (net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr)) dev =dev_get_by_index_rcu (net,inet_iif (skb_in));if (dev)
net/ipv4/sysctl_net_ipv4.c +2 -0
@@ -639,6 +639,8 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 =SYSCTL_ONE }, { .procname = "icmp_ratelimit",
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 117/231 icmp: Fix a data-race around sysctl_icmp_ratelimit.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 2a4eb714841f288cf51c7d942d98af6a8c6e4b01 ]
While reading sysctl_icmp_ratelimit, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/icmp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/ipv4/icmp.c +2 -1
@@ -320,7 +320,8 @@ static bool icmpv4_xrlim_allow(struct net *net, struct rtable *rt, vif =l3mdev_master_ifindex (dst->dev); peer =inet_getpeer_v4 (net->ipv4.peers, fl4->daddr, vif,1 ); - rc =inet_peer_xrlim_allow (peer, net->ipv4.sysctl_icmp_ratelimit); + rc =inet_peer_xrlim_allow (peer, +READ_ONCE (net->ipv4.sysctl_icmp_ratelimit));if (peer)inet_putpeer (peer); out:
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 118/231 icmp: Fix a data-race around sysctl_icmp_ratemask.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 1ebcb25ad6fc3d50fca87350acf451b9a66dd31e ]
While reading sysctl_icmp_ratemask, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/icmp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/icmp.c +1 -1
@@ -282,7 +282,7 @@ static bool icmpv4_mask_allow(struct net *net, int type, int code)return true;/* Limit if icmp type is enabled in ratemask. */ -if (!((1 << type) & net->ipv4.sysctl_icmp_ratemask)) +if (!((1 << type) &READ_ONCE (net->ipv4.sysctl_icmp_ratemask)))return true;return false;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 119/231 raw: Fix a data-race around sysctl_raw_l3mdev_accept.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 1dace014928e6e385363032d359a04dee9158af0 ]
While reading sysctl_raw_l3mdev_accept, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- include/net/raw.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/net/raw.h +1 -1
@@ -75,7 +75,7 @@ static inline bool raw_sk_bound_dev_eq(struct net *net, int bound_dev_if, int dif, int sdif) {#if IS_ENABLED (CONFIG_NET_L3_MASTER_DEV ) -return inet_bound_dev_eq (!!net->ipv4.sysctl_raw_l3mdev_accept, +return inet_bound_dev_eq (READ_ONCE (net->ipv4.sysctl_raw_l3mdev_accept), bound_dev_if, dif, sdif);#else return inet_bound_dev_eq (true, bound_dev_if, dif, sdif);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 120/231 tcp: Fix data-races around sysctl_tcp_ecn.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 4785a66702f086cf2ea84bdbe6ec921f274bd9f2 ]
While reading sysctl_tcp_ecn, it can be changed concurrently. Thus, we need to add READ_ONCE() to its readers.
--- drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c | 2 +- net/ipv4/syncookies.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 2 ++ net/ipv4/tcp_input.c | 2 +- net/ipv4/tcp_output.c | 2 +- 5 files changed, 6 insertions(+), 4 deletions(-)
drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c +1 -1
@@ -1392,7 +1392,7 @@ static void chtls_pass_accept_request(struct sock *sk, th_ecn = tcph->ece && tcph->cwr;if (th_ecn) { ect = !INET_ECN_is_not_ect (ip_dsfield); - ecn_ok =sock_net (sk)->ipv4.sysctl_tcp_ecn; + ecn_ok =READ_ONCE (sock_net (sk)->ipv4.sysctl_tcp_ecn);if ((!ect && ecn_ok) ||tcp_ca_needs_ecn (sk))inet_rsk (oreq)->ecn_ok =1 ; }
net/ipv4/syncookies.c +1 -1
@@ -273,7 +273,7 @@ bool cookie_ecn_ok(const struct tcp_options_received *tcp_opt,if (!ecn_ok)return false; -if (net->ipv4.sysctl_tcp_ecn) +if (READ_ONCE (net->ipv4.sysctl_tcp_ecn))return true;return dst_feature (dst,RTAX_FEATURE_ECN );
net/ipv4/sysctl_net_ipv4.c +2 -0
@@ -680,6 +680,8 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_TWO, }, { .procname = "tcp_ecn_fallback",
net/ipv4/tcp_input.c +1 -1
@@ -6712,7 +6712,7 @@ static void tcp_ecn_create_request(struct request_sock *req, ect = !INET_ECN_is_not_ect (TCP_SKB_CB (skb)->ip_dsfield); ecn_ok_dst =dst_feature (dst,DST_FEATURE_ECN_MASK ); - ecn_ok = net->ipv4.sysctl_tcp_ecn || ecn_ok_dst; + ecn_ok =READ_ONCE (net->ipv4.sysctl_tcp_ecn) || ecn_ok_dst; if (((!ect || th->res1) && ecn_ok) || tcp_ca_needs_ecn(listen_sk) || (ecn_ok_dst & DST_FEATURE_ECN_CA) ||
net/ipv4/tcp_output.c +1 -1
@@ -324,7 +324,7 @@ static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb) {struct tcp_sock *tp =tcp_sk (sk);bool bpf_needs_ecn =tcp_bpf_ca_needs_ecn (sk); -bool use_ecn =sock_net (sk)->ipv4.sysctl_tcp_ecn ==1 || +bool use_ecn =READ_ONCE (sock_net (sk)->ipv4.sysctl_tcp_ecn) ==1 ||tcp_ca_needs_ecn (sk) || bpf_needs_ecn;if (!use_ecn) {
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 121/231 tcp: Fix a data-race around sysctl_tcp_ecn_fallback.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 12b8d9ca7e678abc48195294494f1815b555d658 ]
While reading sysctl_tcp_ecn_fallback, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.
--- net/ipv4/sysctl_net_ipv4.c | 2 ++ net/ipv4/tcp_output.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-)
net/ipv4/sysctl_net_ipv4.c +2 -0
@@ -689,6 +689,8 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, }, { .procname = "ip_dynaddr",
net/ipv4/tcp_output.c +1 -1
@@ -346,7 +346,7 @@ static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb)static void tcp_ecn_clear_syn (struct sock *sk,struct sk_buff *skb) { -if (sock_net (sk)->ipv4.sysctl_tcp_ecn_fallback) +if (READ_ONCE (sock_net (sk)->ipv4.sysctl_tcp_ecn_fallback))/* tp->ecn_flags are cleared at a later point in time when * SYN ACK is ultimatively being received. */
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 122/231 ipv4: Fix data-races around sysctl_ip_dynaddr.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit e49e4aff7ec19b2d0d0957ee30e93dade57dab9e ]
While reading sysctl_ip_dynaddr, it can be changed concurrently. Thus, we need to add READ_ONCE() to its readers.
--- Documentation/networking/ip-sysctl.rst | 2 +- net/ipv4/af_inet.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)
Documentation/networking/ip-sysctl.rst +1 -1
@@ -1179,7 +1179,7 @@ ip_autobind_reuse - BOOLEAN option should only be set by experts. Default: 0 -ip_dynaddr - BOOLEAN +ip_dynaddr - INTEGER If set non-zero, enables support for dynamic addresses. If set to a non-zero value larger than 1, a kernel log message will be printed when dynamic address rewriting
net/ipv4/af_inet.c +2 -2
@@ -1247,7 +1247,7 @@ static int inet_sk_reselect_saddr(struct sock *sk)if (new_saddr == old_saddr)return 0 ; -if (sock_net (sk)->ipv4.sysctl_ip_dynaddr >1 ) { +if (READ_ONCE (sock_net (sk)->ipv4.sysctl_ip_dynaddr) >1 ) {pr_info ("%s(): shifting inet->saddr from %pI4 to %pI4\n" , __func__, &old_saddr, &new_saddr); }
@@ -1302,7 +1302,7 @@ int inet_sk_rebuild_header(struct sock *sk)* Other protocols have to map its equivalent state to TCP_SYN_SENT. * DCCP maps its DCCP_REQUESTING state to TCP_SYN_SENT. -acme */ -if (!sock_net (sk)->ipv4.sysctl_ip_dynaddr || +if (!READ_ONCE (sock_net (sk)->ipv4.sysctl_ip_dynaddr) || sk->sk_state !=TCP_SYN_SENT || (sk->sk_userlocks &SOCK_BINDADDR_LOCK ) || (err =inet_sk_reselect_saddr (sk)) !=0 )
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 123/231 nexthop: Fix data-races around nexthop_compat_mode.
In reply to Greg Kroah-HartmanFrom: Kuniyuki Iwashima <[email protected]>
[ Upstream commit bdf00bf24bef9be1ca641a6390fd5487873e0d2e ]
While reading nexthop_compat_mode, it can be changed concurrently. Thus, we need to add READ_ONCE() to its readers.
--- net/ipv4/fib_semantics.c | 2 +- net/ipv4/nexthop.c | 5 +++-- net/ipv6/route.c | 2 +- 3 files changed, 5 insertions(+), 4 deletions(-)
net/ipv4/fib_semantics.c +1 -1
@@ -1811,7 +1811,7 @@ int fib_dump_info(struct sk_buff *skb, u goto nla_put_failure;if (nexthop_is_blackhole (fi->nh)) rtm->rtm_type =RTN_BLACKHOLE ; -if (!fi->fib_net->ipv4.sysctl_nexthop_compat_mode) +if (!READ_ONCE (fi->fib_net->ipv4.sysctl_nexthop_compat_mode)) goto offload; }
net/ipv4/nexthop.c +3 -2
@@ -1858,7 +1858,7 @@ static void __remove_nexthop_fib(struct/* __ip6_del_rt does a release, so do a hold here */ fib6_info_hold (f6i); ipv6_stub->ip6_del_rt (net, f6i, - !net->ipv4.sysctl_nexthop_compat_mode); + !READ_ONCE (net->ipv4.sysctl_nexthop_compat_mode)); } }
@@ -2361,7 +2361,8 @@ out:if (!rc) {nh_base_seq_inc (net);nexthop_notify (RTM_NEWNEXTHOP , new_nh, &cfg->nlinfo); -if (replace_notify && net->ipv4.sysctl_nexthop_compat_mode) +if (replace_notify && +READ_ONCE (net->ipv4.sysctl_nexthop_compat_mode))nexthop_replace_notify (net, new_nh, &cfg->nlinfo); }
net/ipv6/route.c +1 -1
@@ -5737,7 +5737,7 @@ static int rt6_fill_node(struct net *netif (nexthop_is_blackhole (rt->nh)) rtm->rtm_type =RTN_BLACKHOLE ; -if (net->ipv4.sysctl_nexthop_compat_mode && +if (READ_ONCE (net->ipv4.sysctl_nexthop_compat_mode) &&rt6_fill_node_nexthop (skb, rt->nh, &nh_flags) <0 ) goto nla_put_failure;
Greg Kroah-Hartman
maintainerPATCH 124/231 net: ftgmac100: Hold reference returned by of_get_child_by_name()
In reply to Greg Kroah-HartmanFrom: Liang He <[email protected]>
[ Upstream commit 49b9f431ff0d845a36be0b3ede35ec324f2e5fee ]
In ftgmac100_probe(), we should hold the refernece returned by of_get_child_by_name() and use it to call of_node_put() for reference balance.
--- drivers/net/ethernet/faraday/ftgmac100.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
drivers/net/ethernet/faraday/ftgmac100.c +14 -1
@@ -1764,6 +1764,19 @@ static int ftgmac100_setup_clk(struct ftgmac100 *priv)return rc; } +static bool ftgmac100_has_child_node (struct device_node *np,const char *name) +{ +struct device_node *child_np =of_get_child_by_name (np, name); +bool ret = false; + +if (child_np) { + ret = true; +of_node_put (child_np); + } + +return ret; +} +static int ftgmac100_probe (struct platform_device *pdev) {struct resource *res;
@@ -1883,7 +1896,7 @@ static int ftgmac100_probe(struct platform_device *pdev)/* Display what we found */ phy_attached_info (phy); - }else if (np && !of_get_child_by_name (np,"mdio" )) { + }else if (np && !ftgmac100_has_child_node (np,"mdio" )) {/* Support legacy ASPEED devicetree descriptions that decribe a * MAC with an embedded MDIO controller but have no "mdio" * child node. Automatically scan the MDIO bus for available
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 125/231 net: stmmac: fix leaks in probe
In reply to Greg Kroah-HartmanFrom: Dan Carpenter <[email protected]>
[ Upstream commit 23aa6d5088e3bd65de77c5c307237b9937f8b48a ]
These two error paths should clean up before returning.
--- drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +4 -2
@@ -273,7 +273,8 @@ static int ingenic_mac_probe(struct platform_device *pdev) mac->tx_delay = tx_delay_ps *1000 ; }else {dev_err (&pdev->dev,"Invalid TX clock delay: %dps\n" , tx_delay_ps); -return -EINVAL ; + ret = -EINVAL ; + goto err_remove_config_dt; } }
@@ -283,7 +284,8 @@ static int ingenic_mac_probe(struct platform_device *pdev) mac->rx_delay = rx_delay_ps *1000 ; }else {dev_err (&pdev->dev,"Invalid RX clock delay: %dps\n" , rx_delay_ps); -return -EINVAL ; + ret = -EINVAL ; + goto err_remove_config_dt; } }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 126/231 ima: force signature verification when CONFIG_KEXEC_SIG is configured
In reply to Greg Kroah-HartmanFrom: Coiby Xu <[email protected]>
[ Upstream commit af16df54b89dee72df253abc5e7b5e8a6d16c11c ]
Currently, an unsigned kernel could be kexec'ed when IMA arch specific policy is configured unless lockdown is enabled. Enforce kernel signature verification check in the kexec_file_load syscall when IMA arch specific policy is configured.
--- include/linux/kexec.h | 6 ++++++ kernel/kexec_file.c | 11 ++++++++++- security/integrity/ima/ima_efi.c | 2 ++ 3 files changed, 18 insertions(+), 1 deletion(-)
include/linux/kexec.h +6 -0
@@ -452,6 +452,12 @@ static inline int kexec_crash_loaded(void) { return 0; }#define kexec_in_progress false#endif /* CONFIG_KEXEC_CORE */ +#ifdef CONFIG_KEXEC_SIG +void set_kexec_sig_enforced (void ); +#else +static inline void set_kexec_sig_enforced (void ) {} +#endif +#endif /* !defined(__ASSEBMLY__) */ #endif /* LINUX_KEXEC_H */
kernel/kexec_file.c +10 -1
@@ -29,6 +29,15 @@#include <linux/vmalloc.h> #include "kexec_internal.h" +#ifdef CONFIG_KEXEC_SIG +static bool sig_enforce =IS_ENABLED (CONFIG_KEXEC_SIG_FORCE ); + +void set_kexec_sig_enforced (void ) +{ + sig_enforce = true; +} +#endif +static int kexec_calculate_store_digests (struct kimage *image);/*
@@ -159,7 +168,7 @@ kimage_validate_signature(struct kimage *image) image->kernel_buf_len);if (ret) { -if (IS_ENABLED (CONFIG_KEXEC_SIG_FORCE )) { +if (sig_enforce) {pr_notice ("Enforced kernel signature verification failed (%d).\n" , ret);return ret; }
security/integrity/ima/ima_efi.c +2 -0
@@ -67,6 +67,8 @@ const char * const *arch_get_ima_policy(void)if (IS_ENABLED (CONFIG_IMA_ARCH_POLICY ) &&arch_ima_get_secureboot ()) {if (IS_ENABLED (CONFIG_MODULE_SIG ))set_module_sig_enforced (); +if (IS_ENABLED (CONFIG_KEXEC_SIG )) +set_kexec_sig_enforced ();return sb_arch_rules; }return NULL ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 127/231 ima: Fix potential memory leak in ima_init_crypto()
In reply to Greg Kroah-HartmanFrom: Jianglei Nie <[email protected]>
[ Upstream commit 067d2521874135267e681c19d42761c601d503d6 ]
On failure to allocate the SHA1 tfm, IMA fails to initialize and exits without freeing the ima_algo_array. Add the missing kfree() for ima_algo_array to avoid the potential memory leak.
--- security/integrity/ima/ima_crypto.c | 1 + 1 file changed, 1 insertion(+)
security/integrity/ima/ima_crypto.c +1 -0
@@ -205,6 +205,7 @@ int __init ima_init_crypto(void)crypto_free_shash (ima_algo_array[i].tfm); } +kfree (ima_algo_array); out:crypto_free_shash (ima_shash_tfm);return rc;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 128/231 drm/amd/display: Ignore First MST Sideband Message Return Error
In reply to Greg Kroah-HartmanFrom: Fangzhi Zuo <[email protected]>
[ Upstream commit acea108fa067d140bd155161a79b1fcd967f4137 ]
[why]
First MST sideband message returns AUX_RET_ERROR_HPD_DISCON on certain intel platform. Aux transaction considered failure if HPD unexpected pulled low. The actual aux transaction success in such case, hence do not return error.
[how]
Not returning error when AUX_RET_ERROR_HPD_DISCON detected on the first sideband message.
v2: squash in additional DMI entries
v3: squash in static fix
--- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 39 +++++++++++++++++++ .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 8 ++++ .../display/amdgpu_dm/amdgpu_dm_mst_types.c | 17 ++++++++ 3 files changed, 64 insertions(+)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +39 -0
@@ -72,6 +72,7 @@#include <linux/pci.h> #include <linux/firmware.h> #include <linux/component.h> +#include <linux/dmi.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_uapi.h>
@@ -1391,6 +1392,41 @@ static bool dm_should_disable_stutter(struct pci_dev *pdev)return false; } +static const struct dmi_system_id hpd_disconnect_quirk_table[] = { + { + .matches = { +DMI_MATCH (DMI_SYS_VENDOR ,"Dell Inc." ), +DMI_MATCH (DMI_PRODUCT_NAME ,"Precision 3660" ), + }, + }, + { + .matches = { +DMI_MATCH (DMI_SYS_VENDOR ,"Dell Inc." ), +DMI_MATCH (DMI_PRODUCT_NAME ,"Precision 3260" ), + }, + }, + { + .matches = { +DMI_MATCH (DMI_SYS_VENDOR ,"Dell Inc." ), +DMI_MATCH (DMI_PRODUCT_NAME ,"Precision 3460" ), + }, + }, + {} +}; + +static void retrieve_dmi_info (struct amdgpu_display_manager *dm) +{ +const struct dmi_system_id *dmi_id; + + dm->aux_hpd_discon_quirk = false; + + dmi_id =dmi_first_match (hpd_disconnect_quirk_table); +if (dmi_id) { + dm->aux_hpd_discon_quirk = true; +DRM_INFO ("aux_hpd_discon_quirk attached\n" ); + } +} +static int amdgpu_dm_init (struct amdgpu_device *adev) {struct dc_init_data init_data;
@@ -1521,6 +1557,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) }INIT_LIST_HEAD (&adev->dm.da_list); + +retrieve_dmi_info (&adev->dm); +/* Display Core create. */ adev->dm.dc =dc_create (&init_data);
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +8 -0
@@ -546,6 +546,14 @@ struct amdgpu_display_manager {* last successfully applied backlight values. */ u32 actual_brightness[AMDGPU_DM_MAX_NUM_EDP ]; + +/** +* @aux_hpd_discon_quirk: +* +* quirk for hpd discon while aux is on-going. +* occurred on certain intel platform +*/ +bool aux_hpd_discon_quirk; };enum dsc_clock_force_state {
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +17 -0
@@ -58,6 +58,8 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,ssize_t result =0 ;struct aux_payload payload;enum aux_return_code_type operation_result; +struct amdgpu_device *adev; +struct ddc_service *ddc;if (WARN_ON (msg->size >16 ))return -E2BIG ;
@@ -76,6 +78,21 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, result =dc_link_aux_transfer_raw (TO_DM_AUX (aux)->ddc_service, &payload, &operation_result); +/* +* w/a on certain intel platform where hpd is unexpected to pull low during +* 1st sideband message transaction by return AUX_RET_ERROR_HPD_DISCON +* aux transaction is succuess in such case, therefore bypass the error +*/ + ddc =TO_DM_AUX (aux)->ddc_service; + adev = ddc->ctx->driver_context; +if (adev->dm.aux_hpd_discon_quirk) { +if (msg->address ==DP_SIDEBAND_MSG_DOWN_REQ_BASE && + operation_result ==AUX_RET_ERROR_HPD_DISCON ) { + result =0 ; + operation_result =AUX_RET_SUCCESS ; + } + } +if (payload.write && result >=0 ) result = msg->size;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 129/231 drm/amdkfd: correct the MEC atomic support firmware checking for GC 10.3.7
In reply to Greg Kroah-HartmanFrom: Prike Liang <[email protected]>
[ Upstream commit c0044865480a162146b9dfe7783e73a08e97b2b9 ]
On the GC 10.3.7 platform the initial MEC release version #3 can support atomic operation,so need correct and set its MEC atomic support version to #3.
--- drivers/gpu/drm/amd/amdkfd/kfd_device.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/gpu/drm/amd/amdkfd/kfd_device.c +2 -0
@@ -158,6 +158,8 @@ static void kfd_device_info_init(struct kfd_dev *kfd,/* Navi2x+, Navi1x+ */ if (gc_version ==IP_VERSION (10 ,3 ,6 )) kfd->device_info.no_atomic_fw_version =14 ; +else if (gc_version ==IP_VERSION (10 ,3 ,7 )) + kfd->device_info.no_atomic_fw_version =3 ;else if (gc_version >=IP_VERSION (10 ,3 ,0 )) kfd->device_info.no_atomic_fw_version =92 ;else if (gc_version >=IP_VERSION (10 ,1 ,1 ))
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 130/231 drm/amd/display: Only use depth 36 bpp linebuffers on DCN display engines.
In reply to Greg Kroah-HartmanFrom: Mario Kleiner <[email protected]>
[ Upstream commit add61d3c31de6a4b5e11a2ab96aaf4c873481568 ]
Various DCE versions had trouble with 36 bpp lb depth, requiring fixes, last time in commit 353ca0fa5630 ("drm/amd/display: Fix 10bit 4K display on CIK GPUs") for DCE-8. So far >= DCE-11.2 was considered ok, but now I found out that on DCE-11.2 it causes dithering when there shouldn't be any, so identity pixel passthrough with identity gamma LUTs doesn't work when it should. This breaks various important neuroscience applications, as reported to me by scientific users of Polaris cards under Ubuntu 22.04 with Linux 5.15, and confirmed by testing it myself on DCE-11.2.
Lets only use depth 36 for DCN engines, where my testing showed that it is both necessary for high color precision output, e.g., RGBA16 fb's, and not harmful, as far as more than one year in real-world use showed.
DCE engines seem to work fine for high precision output at 30 bpp, so this ("famous last words") depth 30 should hopefully fix all known problems without introducing new ones.
Successfully retested on DCE-11.2 Polaris and DCN-1.0 Raven Ridge on top of Linux 5.19.0-rc2 + drm-next.
--- drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/display/dc/core/dc_resource.c +6 -5
@@ -1113,12 +1113,13 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx)* on certain displays, such as the Sharp 4k. 36bpp is needed * to support SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616 and * SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616 with actual > 10 bpc -* precision on at least DCN display engines. However, at least -* Carrizo with DCE_VERSION_11_0 does not like 36 bpp lb depth, -* so use only 30 bpp on DCE_VERSION_11_0. Testing with DCE 11.2 and 8.3 -* did not show such problems, so this seems to be the exception. +* precision on DCN display engines, but apparently not for DCE, as +* far as testing on DCE-11.2 and DCE-8 showed. Various DCE parts have +* problems: Carrizo with DCE_VERSION_11_0 does not like 36 bpp lb depth, +* neither do DCE-8 at 4k resolution, or DCE-11.2 (broken identify pixel +* passthrough). Therefore only use 36 bpp on DCN where it is actually needed. */ -if (plane_state->ctx->dce_version >DCE_VERSION_11_0 ) +if (plane_state->ctx->dce_version >DCE_VERSION_MAX ) pipe_ctx->plane_res.scl_data.lb_params.depth =LB_PIXEL_DEPTH_36BPP ;else pipe_ctx->plane_res.scl_data.lb_params.depth =LB_PIXEL_DEPTH_30BPP ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 131/231 drm/amd/pm: Prevent divide by zero
In reply to Greg Kroah-HartmanFrom: Yefim Barashkin <[email protected]>
[ Upstream commit 0638c98c17aa12fe914459c82cd178247e21fb2b ]
divide error: 0000 [#1] SMP PTI
CPU: 3 PID: 78925 Comm: tee Not tainted 5.15.50-1-lts #1 Hardware name: MSI MS-7A59/Z270 SLI PLUS (MS-7A59), BIOS 1.90 01/30/2018 RIP: 0010:smu_v11_0_set_fan_speed_rpm+0x11/0x110 [amdgpu]
Speed is user-configurable through a file.
I accidentally set it to zero, and the driver crashed.
--- drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c +2 -0
@@ -1228,6 +1228,8 @@ int smu_v11_0_set_fan_speed_rpm(struct smu_context *smu,uint32_t crystal_clock_freq =2500 ;uint32_t tach_period; +if (speed ==0 ) +return -EINVAL ;/* * To prevent from possible overheat, some ASICs may have requirement * for minimum fan speed:
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 132/231 drm/amd/display: Ensure valid event timestamp for cursor-only commits
In reply to Greg Kroah-HartmanFrom: Michel Dänzer <[email protected]>
[ Upstream commit 3283c83eb6fcfbda8ea03d7149d8e42e71c5d45e ]
Requires enabling the vblank machinery for them.
--- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 43 ++++++++++++++++++++-- 1 file changed, 40 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +40 -3
@@ -464,6 +464,26 @@ static void dm_pflip_high_irq(void *inte vrr_active, (int) !e); } +static void dm_crtc_handle_vblank (struct amdgpu_crtc *acrtc) +{ +struct drm_crtc *crtc = &acrtc->base; +struct drm_device *dev = crtc->dev; +unsigned long flags; + +drm_crtc_handle_vblank (crtc); + +spin_lock_irqsave (&dev->event_lock, flags); + +/* Send completion event for cursor-only commits */ +if (acrtc->event && acrtc->pflip_status !=AMDGPU_FLIP_SUBMITTED ) { +drm_crtc_send_vblank_event (crtc, acrtc->event); +drm_crtc_vblank_put (crtc); + acrtc->event =NULL ; + } + +spin_unlock_irqrestore (&dev->event_lock, flags); +} +static void dm_vupdate_high_irq (void *interrupt_params) {struct common_irq_params *irq_params = interrupt_params;
@@ -502,7 +522,7 @@ static void dm_vupdate_high_irq(void *in* if a pageflip happened inside front-porch. */ if (vrr_active) { -drm_crtc_handle_vblank (&acrtc->base); +dm_crtc_handle_vblank (acrtc);/* BTR processing for pre-DCE12 ASICs */ if (acrtc->dm_irq_params.stream &&
@@ -554,7 +574,7 @@ static void dm_crtc_high_irq(void *inter* to dm_vupdate_high_irq after end of front-porch. */ if (!vrr_active) -drm_crtc_handle_vblank (&acrtc->base); +dm_crtc_handle_vblank (acrtc);/** * Following stuff must happen at start of vblank, for crc
@@ -9199,6 +9219,7 @@ static void amdgpu_dm_commit_planes(strustruct amdgpu_bo *abo;uint32_t target_vblank, last_flip_vblank;bool vrr_active =amdgpu_dm_vrr_active (acrtc_state); +bool cursor_update = false;bool pflip_present = false;struct {struct dc_surface_update surface_updates[MAX_SURFACES ];
@@ -9234,8 +9255,13 @@ static void amdgpu_dm_commit_planes(strustruct dm_plane_state *dm_new_plane_state =to_dm_plane_state (new_plane_state);/* Cursor plane is handled after stream updates */ -if (plane->type ==DRM_PLANE_TYPE_CURSOR ) +if (plane->type ==DRM_PLANE_TYPE_CURSOR ) { +if ((fb && crtc == pcrtc) || + (old_plane_state->fb && old_plane_state->crtc == pcrtc)) + cursor_update = true; +continue ; + }if (!fb || !crtc || pcrtc != crtc)continue ;
@@ -9397,6 +9423,17 @@ static void amdgpu_dm_commit_planes(stru bundle->stream_update.vrr_infopacket = &acrtc_state->stream->vrr_infopacket; } + }else if (cursor_update && acrtc_state->active_planes >0 && + !acrtc_state->force_dpms_off && + acrtc_attach->base.state->event) { +drm_crtc_vblank_get (pcrtc); + +spin_lock_irqsave (&pcrtc->dev->event_lock, flags); + + acrtc_attach->event = acrtc_attach->base.state->event; + acrtc_attach->base.state->event =NULL ; + +spin_unlock_irqrestore (&pcrtc->dev->event_lock, flags); }/* Update the planes if changed or disable if we don't have any. */
Greg Kroah-Hartman
maintainerPATCH 133/231 smb3: workaround negprot bug in some Samba servers
In reply to Greg Kroah-HartmanFrom: Steve French <[email protected]>
[ Upstream commit 32f319183c439b239294cb2d70ada3564c4c7c39 ]
Mount can now fail to older Samba servers due to a server bug handling padding at the end of the last negotiate context (negotiate contexts typically are rounded up to 8 bytes by adding padding if needed). This server bug can be avoided by switching the order of negotiate contexts, placing a negotiate context at the end that does not require padding (prior to the recent netname context fix this was the case on the client).
--- fs/cifs/smb2pdu.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
fs/cifs/smb2pdu.c +7 -6
@@ -571,10 +571,6 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, *total_len += ctxt_len; pneg_ctxt += ctxt_len; -build_posix_ctxt ((struct smb2_posix_neg_context *)pneg_ctxt); - *total_len +=sizeof (struct smb2_posix_neg_context ); - pneg_ctxt +=sizeof (struct smb2_posix_neg_context ); -/* * secondary channels don't have the hostname field populated * use the hostname field in the primary channel instead
@@ -586,9 +582,14 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, hostname); *total_len += ctxt_len; pneg_ctxt += ctxt_len; - neg_context_count =4 ; - }else /* second channels do not have a hostname */ neg_context_count =3 ; + }else + neg_context_count =2 ; + +build_posix_ctxt ((struct smb2_posix_neg_context *)pneg_ctxt); + *total_len +=sizeof (struct smb2_posix_neg_context ); + pneg_ctxt +=sizeof (struct smb2_posix_neg_context ); + neg_context_count++;if (server->compress_algorithm) { build_compression_ctxt((struct smb2_compression_capabilities_context *)
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 134/231 sfc: fix use after free when disabling sriov
In reply to Greg Kroah-HartmanFrom: Íñigo Huguet <[email protected]>
[ Upstream commit ebe41da5d47ac0fff877e57bd14c54dccf168827 ]
Use after free is detected by kfence when disabling sriov. What was read after being freed was vf->pci_dev: it was freed from pci_disable_sriov and later read in efx_ef10_sriov_free_vf_vports, called from efx_ef10_sriov_free_vf_vswitching.
Set the pointer to NULL at release time to not trying to read it later.
Reproducer and dmesg log (note that kfence doesn't detect it every time):
$ echo 1 > /sys/class/net/enp65s0f0np0/device/sriov_numvfs $ echo 0 > /sys/class/net/enp65s0f0np0/device/sriov_numvfs
BUG: KFENCE: use-after-free read in efx_ef10_sriov_free_vf_vswitching+0x82/0x170 [sfc]
Use-after-free read at 0x00000000ff3c1ba5 (in kfence-#224):
efx_ef10_sriov_free_vf_vswitching+0x82/0x170 [sfc] efx_ef10_pci_sriov_disable+0x38/0x70 [sfc] efx_pci_sriov_configure+0x24/0x40 [sfc] sriov_numvfs_store+0xfe/0x140 kernfs_fop_write_iter+0x11c/0x1b0 new_sync_write+0x11f/0x1b0 vfs_write+0x1eb/0x280 ksys_write+0x5f/0xe0 do_syscall_64+0x5c/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xae
kfence-#224: 0x00000000edb8ef95-0x00000000671f5ce1, size=2792, cache=kmalloc-4k
allocated by task 6771 on cpu 10 at 3137.860196s:
pci_alloc_dev+0x21/0x60 pci_iov_add_virtfn+0x2a2/0x320 sriov_enable+0x212/0x3e0 efx_ef10_sriov_configure+0x67/0x80 [sfc] efx_pci_sriov_configure+0x24/0x40 [sfc] sriov_numvfs_store+0xba/0x140 kernfs_fop_write_iter+0x11c/0x1b0 new_sync_write+0x11f/0x1b0 vfs_write+0x1eb/0x280 ksys_write+0x5f/0xe0 do_syscall_64+0x5c/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xae
freed by task 6771 on cpu 12 at 3170.991309s:
device_release+0x34/0x90 kobject_cleanup+0x3a/0x130 pci_iov_remove_virtfn+0xd9/0x120 sriov_disable+0x30/0xe0 efx_ef10_pci_sriov_disable+0x57/0x70 [sfc] efx_pci_sriov_configure+0x24/0x40 [sfc] sriov_numvfs_store+0xfe/0x140 kernfs_fop_write_iter+0x11c/0x1b0 new_sync_write+0x11f/0x1b0 vfs_write+0x1eb/0x280 ksys_write+0x5f/0xe0 do_syscall_64+0x5c/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xae
--- drivers/net/ethernet/sfc/ef10_sriov.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/net/ethernet/sfc/ef10_sriov.c +7 -3
@@ -408,8 +408,9 @@ static int efx_ef10_pci_sriov_enable(struct efx_nic *efx, int num_vfs)static int efx_ef10_pci_sriov_disable (struct efx_nic *efx,bool force) {struct pci_dev *dev = efx->pci_dev; +struct efx_ef10_nic_data *nic_data = efx->nic_data;unsigned int vfs_assigned =pci_vfs_assigned (dev); -int rc =0 ; +int i, rc =0 ;if (vfs_assigned && !force) { netif_info(efx, drv, efx->net_dev, "VFs are assigned to guests; "
@@ -417,10 +418,13 @@ static int efx_ef10_pci_sriov_disable(struct efx_nic *efx, bool force)return -EBUSY ; } -if (!vfs_assigned) +if (!vfs_assigned) { +for (i =0 ; i < efx->vf_count; i++) + nic_data->vf[i].pci_dev =NULL ;pci_disable_sriov (dev); -else + }else { rc = -EBUSY ; + }efx_ef10_sriov_free_vf_vswitching (efx); efx->vf_count =0 ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 135/231 netfs: do not unlock and put the folio twice
In reply to Greg Kroah-HartmanFrom: Xiubo Li <[email protected]>
[ Upstream commit fac47b43c760ea90e64b895dba60df0327be7775 ]
check_write_begin() will unlock and put the folio when return non-zero. So we should avoid unlocking and putting it twice in netfs layer.
Change the way ->check_write_begin() works in the following two ways:
(1) Pass it a pointer to the folio pointer, allowing it to unlock and put the folio prior to doing the stuff it wants to do, provided it clears the folio pointer.
(2) Change the return values such that 0 with folio pointer set means continue, 0 with folio pointer cleared means re-get and all error codes indicating an error (no special treatment for -EAGAIN).
[ bagasdotme: use Sphinx code text syntax for *foliop pointer ]
--- Documentation/filesystems/netfs_library.rst | 8 +++++--- fs/afs/file.c | 2 +- fs/ceph/addr.c | 11 ++++++----- fs/netfs/buffered_read.c | 17 ++++++++++------- include/linux/netfs.h | 2 +- 5 files changed, 23 insertions(+), 17 deletions(-)
Documentation/filesystems/netfs_library.rst +5 -3
@@ -300,7 +300,7 @@ through which it can issue requests and negotiate:: void (*issue_read)(struct netfs_io_subrequest *subreq); bool (*is_still_valid)(struct netfs_io_request *rreq); int (*check_write_begin)(struct file *file, loff_t pos, unsigned len, - struct folio *folio, void **_fsdata); + struct folio **foliop, void **_fsdata); void (*done)(struct netfs_io_request *rreq); void (*cleanup)(struct address_space *mapping, void *netfs_priv); };
@@ -376,8 +376,10 @@ The operations are as follows: allocated/grabbed the folio to be modified to allow the filesystem to flush conflicting state before allowing it to be modified. - It should return 0 if everything is now fine, -EAGAIN if the folio should be - regrabbed and any other error code to abort the operation. + It may unlock and discard the folio it was given and set the caller's folio + pointer to NULL. It should return 0 if everything is now fine (``*foliop`` + left set) or the op should be retried (``*foliop`` cleared) and any other + error code to abort the operation. * ``done``
fs/afs/file.c +1 -1
@@ -376,7 +376,7 @@ static int afs_begin_cache_operation(struct netfs_io_request *rreq) }static int afs_check_write_begin (struct file *file,loff_t pos,unsigned len, -struct folio *folio,void **_fsdata) +struct folio **foliop,void **_fsdata) {struct afs_vnode *vnode =AFS_FS_I (file_inode (file));
fs/ceph/addr.c +6 -5
@@ -63,7 +63,7 @@ (CONGESTION_ON_THRESH(congestion_kb) >> 2))static int ceph_netfs_check_write_begin (struct file *file,loff_t pos,unsigned int len, -struct folio *folio,void **_fsdata); +struct folio **foliop,void **_fsdata);static inline struct ceph_snap_context *page_snap_context (struct page *page) {
@@ -1285,18 +1285,19 @@ ceph_find_incompatible(struct page *page) }static int ceph_netfs_check_write_begin (struct file *file,loff_t pos,unsigned int len, -struct folio *folio,void **_fsdata) +struct folio **foliop,void **_fsdata) {struct inode *inode =file_inode (file);struct ceph_inode_info *ci =ceph_inode (inode);struct ceph_snap_context *snapc; - snapc =ceph_find_incompatible (folio_page (folio,0 )); + snapc =ceph_find_incompatible (folio_page (*foliop,0 ));if (snapc) {int r; -folio_unlock (folio); -folio_put (folio); +folio_unlock (*foliop); +folio_put (*foliop); + *foliop =NULL ;if (IS_ERR (snapc))return PTR_ERR (snapc);
fs/netfs/buffered_read.c +10 -7
@@ -320,8 +320,9 @@ static bool netfs_skip_folio_read(struct folio *folio, loff_t pos, size_t len,* conflicting writes once the folio is grabbed and locked. It is passed a * pointer to the fsdata cookie that gets returned to the VM to be passed to * write_end. It is permitted to sleep. It should return 0 if the request -* should go ahead; unlock the folio and return -EAGAIN to cause the folio to -* be regot; or return an error. +* should go ahead or it may return an error. It may also unlock and put the +* folio, provided it sets ``*foliop`` to NULL, in which case a return of 0 +* will cause the folio to be re-got and the process to be retried. * * The calling netfs must initialise a netfs context contiguous to the vfs * inode before calling this.
@@ -352,13 +353,13 @@ int netfs_write_begin(struct file *file, struct address_space *mapping,if (ctx->ops->check_write_begin) {/* Allow the netfs (eg. ceph) to flush conflicts. */ - ret = ctx->ops->check_write_begin (file, pos, len, folio, _fsdata); + ret = ctx->ops->check_write_begin (file, pos, len, &folio, _fsdata);if (ret <0 ) {trace_netfs_failure (NULL ,NULL , ret, netfs_fail_check_write_begin); -if (ret == -EAGAIN ) - goto retry; goto error; } +if (!folio) + goto retry; }if (folio_test_uptodate (folio))
@@ -420,8 +421,10 @@ int netfs_write_begin(struct file *file, struct address_space *mapping, error_put:netfs_put_request (rreq, false, netfs_rreq_trace_put_failed); error: -folio_unlock (folio); -folio_put (folio); +if (folio) { +folio_unlock (folio); +folio_put (folio); + }_leave (" = %d" , ret);return ret; }
include/linux/netfs.h +1 -1
@@ -211,7 +211,7 @@ struct netfs_request_ops {void (*issue_read)(struct netfs_io_subrequest *subreq);bool (*is_still_valid)(struct netfs_io_request *rreq);int (*check_write_begin)(struct file *file,loff_t pos,unsigned len, -struct folio *folio,void **_fsdata); +struct folio **foliop,void **_fsdata);void (*done)(struct netfs_io_request *rreq);void (*cleanup)(struct address_space *mapping,void *netfs_priv); };
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 136/231 seg6: fix skb checksum evaluation in SRH encapsulation/insertion
In reply to Greg Kroah-HartmanFrom: Andrea Mayer <[email protected]>
[ Upstream commit df8386d13ea280d55beee1b95f61a59234a3798b ]
Support for SRH encapsulation and insertion was introduced with commit 6c8702c60b88 ("ipv6: sr: add support for SRH encapsulation and injection with lwtunnels"), through the seg6_do_srh_encap() and seg6_do_srh_inline() functions, respectively.
The former encapsulates the packet in an outer IPv6 header along with the SRH, while the latter inserts the SRH between the IPv6 header and the payload. Then, the headers are initialized/updated according to the operating mode (i.e., encap/inline).
Finally, the skb checksum is calculated to reflect the changes applied to the headers.
The IPv6 payload length ('payload_len') is not initialized within seg6_do_srh_{inline,encap}() but is deferred in seg6_do_srh(), i.e. the caller of seg6_do_srh_{inline,encap}().
However, this operation invalidates the skb checksum, since the 'payload_len' is updated only after the checksum is evaluated.
To solve this issue, the initialization of the IPv6 payload length is moved from seg6_do_srh() directly into the seg6_do_srh_{inline,encap}() functions and before the skb checksum update takes place.
--- net/ipv6/seg6_iptunnel.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/ipv6/seg6_iptunnel.c +4 -1
@@ -189,6 +189,8 @@ int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto) }#endif + hdr->payload_len =htons (skb->len -sizeof (struct ipv6hdr )); +skb_postpush_rcsum (skb, hdr, tot_len);return 0 ;
@@ -241,6 +243,8 @@ int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) }#endif + hdr->payload_len =htons (skb->len -sizeof (struct ipv6hdr )); +skb_postpush_rcsum (skb, hdr,sizeof (struct ipv6hdr ) + hdrlen);return 0 ;
@@ -302,7 +306,6 @@ static int seg6_do_srh(struct sk_buff *skb)break ; } -ipv6_hdr (skb)->payload_len =htons (skb->len -sizeof (struct ipv6hdr ));skb_set_transport_header (skb,sizeof (struct ipv6hdr ));nf_reset_ct (skb);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 137/231 seg6: fix skb checksum in SRv6 End.B6 and End.B6.Encaps behaviors
In reply to Greg Kroah-HartmanFrom: Andrea Mayer <[email protected]>
[ Upstream commit f048880fc77058d864aff5c674af7918b30f312a ]
The SRv6 End.B6 and End.B6.Encaps behaviors rely on functions seg6_do_srh_{encap,inline}() to, respectively: i) encapsulate the packet within an outer IPv6 header with the specified Segment Routing Header (SRH); ii) insert the specified SRH directly after the IPv6 header of the packet.
This patch removes the initialization of the IPv6 header payload length from the input_action_end_b6{_encap}() functions, as it is now handled properly by seg6_do_srh_{encap,inline}() to avoid corruption of the skb checksum.
--- net/ipv6/seg6_local.c | 2 -- 1 file changed, 2 deletions(-)
net/ipv6/seg6_local.c +0 -2
@@ -826,7 +826,6 @@ static int input_action_end_b6(struct sk_buff *skb, struct seg6_local_lwt *slwt)if (err) goto drop; -ipv6_hdr (skb)->payload_len =htons (skb->len -sizeof (struct ipv6hdr ));skb_set_transport_header (skb,sizeof (struct ipv6hdr ));seg6_lookup_nexthop (skb,NULL ,0 );
@@ -858,7 +857,6 @@ static int input_action_end_b6_encap(struct sk_buff *skb,if (err) goto drop; -ipv6_hdr (skb)->payload_len =htons (skb->len -sizeof (struct ipv6hdr ));skb_set_transport_header (skb,sizeof (struct ipv6hdr ));seg6_lookup_nexthop (skb,NULL ,0 );
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 138/231 seg6: bpf: fix skb checksum in bpf_push_seg6_encap()
In reply to Greg Kroah-HartmanFrom: Andrea Mayer <[email protected]>
[ Upstream commit 4889fbd98deaf243c3baadc54e296d71c6af1eb0 ]
Both helper functions bpf_lwt_seg6_action() and bpf_lwt_push_encap() use the bpf_push_seg6_encap() to encapsulate the packet in an IPv6 with Segment Routing Header (SRH) or insert an SRH between the IPv6 header and the payload.
To achieve this result, such helper functions rely on bpf_push_seg6_encap() which, in turn, leverages seg6_do_srh_{encap,inline}() to perform the required operation (i.e. encap/inline).
This patch removes the initialization of the IPv6 header payload length from bpf_push_seg6_encap(), as it is now handled properly by seg6_do_srh_{encap,inline}() to prevent corruption of the skb checksum.
--- net/core/filter.c | 1 - 1 file changed, 1 deletion(-)
net/core/filter.c +0 -1
@@ -6148,7 +6148,6 @@ static int bpf_push_seg6_encap(struct sk_buff *skb, u32 type, void *hdr, u32 lenif (err)return err; -ipv6_hdr (skb)->payload_len =htons (skb->len -sizeof (struct ipv6hdr ));skb_set_transport_header (skb,sizeof (struct ipv6hdr ));return seg6_lookup_nexthop (skb,NULL ,0 );
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 139/231 sfc: fix kernel panic when creating VF
In reply to Greg Kroah-HartmanFrom: Íñigo Huguet <[email protected]>
[ Upstream commit ada74c5539eba06cf8b47d068f92e0b3963a9a6e ]
When creating VFs a kernel panic can happen when calling to efx_ef10_try_update_nic_stats_vf.
When releasing a DMA coherent buffer, sometimes, I don't know in what specific circumstances, it has to unmap memory with vunmap. It is disallowed to do that in IRQ context or with BH disabled. Otherwise, we hit this line in vunmap, causing the crash:
BUG_ON(in_interrupt());
This patch reenables BH to release the buffer.
Log messages when the bug is hit:
kernel BUG at mm/vmalloc.c:2727!
invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
CPU: 6 PID: 1462 Comm: NetworkManager Kdump: loaded Tainted: G I --------- --- 5.14.0-119.el9.x86_64 #1 Hardware name: Dell Inc. PowerEdge R740/06WXJT, BIOS 2.8.2 08/27/2020 RIP: 0010:vunmap+0x2e/0x30 ...skip... Call Trace: __iommu_dma_free+0x96/0x100 efx_nic_free_buffer+0x2b/0x40 [sfc] efx_ef10_try_update_nic_stats_vf+0x14a/0x1c0 [sfc] efx_ef10_update_stats_vf+0x18/0x40 [sfc] efx_start_all+0x15e/0x1d0 [sfc] efx_net_open+0x5a/0xe0 [sfc] __dev_open+0xe7/0x1a0 __dev_change_flags+0x1d7/0x240 dev_change_flags+0x21/0x60 ...skip...
--- drivers/net/ethernet/sfc/ef10.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/ethernet/sfc/ef10.c +3 -0
@@ -1932,7 +1932,10 @@ static int efx_ef10_try_update_nic_stats_vf(struct efx_nic *efx)efx_update_sw_stats (efx, stats); out: +/* releasing a DMA coherent buffer with BH disabled can panic */ +spin_unlock_bh (&efx->stats_lock);efx_nic_free_buffer (efx, &stats_buf); +spin_lock_bh (&efx->stats_lock);return rc; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 140/231 net: atlantic: remove deep parameter on suspend/resume functions
In reply to Greg Kroah-HartmanFrom: Chia-Lin Kao (AceLan) <[email protected]>
[ Upstream commit 0f33250760384e05c36466b0a2f92f3c6007ba92 ]
Below commit claims that atlantic NIC requires to reset the device on pm
op, and had set the deep to true for all suspend/resume functions.
commit 1809c30b6e5a ("net: atlantic: always deep reset on pm op, fixing up my null deref regression") So, we could remove deep parameter on suspend/resume functions without
any functional change.
--- .../ethernet/aquantia/atlantic/aq_pci_func.c | 24 ++++++++----------- 1 file changed, 10 insertions(+), 14 deletions(-)
drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c +10 -14
@@ -379,7 +379,7 @@ static void aq_pci_shutdown(struct pci_dev *pdev) } } -static int aq_suspend_common (struct device *dev,bool deep) +static int aq_suspend_common (struct device *dev) {struct aq_nic_s *nic =pci_get_drvdata (to_pci_dev (dev));
@@ -392,17 +392,15 @@ static int aq_suspend_common(struct device *dev, bool deep)if (netif_running (nic->ndev))aq_nic_stop (nic); -if (deep) { -aq_nic_deinit (nic, !nic->aq_hw->aq_nic_cfg->wol); -aq_nic_set_power (nic); - } +aq_nic_deinit (nic, !nic->aq_hw->aq_nic_cfg->wol); +aq_nic_set_power (nic);rtnl_unlock ();return 0 ; } -static int atl_resume_common (struct device *dev,bool deep) +static int atl_resume_common (struct device *dev) {struct pci_dev *pdev =to_pci_dev (dev);struct aq_nic_s *nic;
@@ -415,10 +413,8 @@ static int atl_resume_common(struct device *dev, bool deep)pci_set_power_state (pdev,PCI_D0 );pci_restore_state (pdev); -if (deep) { -/* Reinitialize Nic/Vecs objects */ -aq_nic_deinit (nic, !nic->aq_hw->aq_nic_cfg->wol); - } +/* Reinitialize Nic/Vecs objects */ +aq_nic_deinit (nic, !nic->aq_hw->aq_nic_cfg->wol);if (netif_running (nic->ndev)) { ret =aq_nic_init (nic);
@@ -444,22 +440,22 @@ static int atl_resume_common(struct device *dev, bool deep)static int aq_pm_freeze (struct device *dev) { -return aq_suspend_common (dev, true); +return aq_suspend_common (dev); }static int aq_pm_suspend_poweroff (struct device *dev) { -return aq_suspend_common (dev, true); +return aq_suspend_common (dev); }static int aq_pm_thaw (struct device *dev) { -return atl_resume_common (dev, true); +return atl_resume_common (dev); }static int aq_pm_resume_restore (struct device *dev) { -return atl_resume_common (dev, true); +return atl_resume_common (dev); }static const struct dev_pm_ops aq_pm_ops = {
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 141/231 net: atlantic: remove aq_nic_deinit() when resume
In reply to Greg Kroah-HartmanFrom: Chia-Lin Kao (AceLan) <[email protected]>
[ Upstream commit 2e15c51fefaffaf9f72255eaef4fada05055e4c5 ]
aq_nic_deinit() has been called while suspending, so we don't have to call it again on resume.
Actually, call it again leads to another hang issue when resuming from S3.
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992345] Call Trace:
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992346] <TASK>
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992348] aq_nic_deinit+0xb4/0xd0 [atlantic]
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992356] aq_pm_thaw+0x7f/0x100 [atlantic]
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992362] pci_pm_resume+0x5c/0x90
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992366] ? pci_pm_thaw+0x80/0x80
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992368] dpm_run_callback+0x4e/0x120
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992371] device_resume+0xad/0x200
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992373] async_resume+0x1e/0x40
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992374] async_run_entry_fn+0x33/0x120
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992377] process_one_work+0x220/0x3c0
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992380] worker_thread+0x4d/0x3f0
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992382] ? process_one_work+0x3c0/0x3c0
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992384] kthread+0x12a/0x150
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992386] ? set_kthread_struct+0x40/0x40
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992387] ret_from_fork+0x22/0x30
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992391] </TASK>
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992392] ---[ end trace 1ec8c79604ed5e0d ]---
Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992394] PM: dpm_run_callback(): pci_pm_resume+0x0/0x90 returns -110 Jul 8 03:09:44 u-Precision-7865-Tower kernel: [ 5910.992397] atlantic 0000:02:00.0: PM: failed to resume async: error -110
--- drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c | 3 --- 1 file changed, 3 deletions(-)
drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c +0 -3
@@ -413,9 +413,6 @@ static int atl_resume_common(struct device *dev)pci_set_power_state (pdev,PCI_D0 );pci_restore_state (pdev); -/* Reinitialize Nic/Vecs objects */ -aq_nic_deinit (nic, !nic->aq_hw->aq_nic_cfg->wol); -if (netif_running (nic->ndev)) { ret =aq_nic_init (nic);if (ret)
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 142/231 KVM: x86: Fully initialize struct kvm_lapic_irq in kvm_pv_kick_cpu_op()
In reply to Greg Kroah-HartmanFrom: Vitaly Kuznetsov <[email protected]>
[ Upstream commit 8a414f943f8b5f94bbaafdec863d6f3dbef33f8a ]
'vector' and 'trig_mode' fields of 'struct kvm_lapic_irq' are left uninitialized in kvm_pv_kick_cpu_op(). While these fields are normally not needed for APIC_DM_REMRD, they're still referenced by __apic_accept_irq() for trace_kvm_apic_accept_irq(). Fully initialize the structure to avoid consuming random stack memory.
Message-Id: <[email protected]>
--- arch/x86/kvm/x86.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)
arch/x86/kvm/x86.c +10 -8
@@ -9074,15 +9074,17 @@ static int kvm_pv_clock_pairing(struct kvm_vcpu *vcpu, gpa_t paddr,*/ static void kvm_pv_kick_cpu_op (struct kvm *kvm,int apicid) { -struct kvm_lapic_irq lapic_irq; - - lapic_irq.shorthand =APIC_DEST_NOSHORT ; - lapic_irq.dest_mode =APIC_DEST_PHYSICAL ; - lapic_irq.level =0 ; - lapic_irq.dest_id = apicid; - lapic_irq.msi_redir_hint = false; +/* +* All other fields are unused for APIC_DM_REMRD, but may be consumed by +* common code, e.g. for tracing. Defer initialization to the compiler. +*/ +struct kvm_lapic_irq lapic_irq = { + .delivery_mode =APIC_DM_REMRD , + .dest_mode =APIC_DEST_PHYSICAL , + .shorthand =APIC_DEST_NOSHORT , + .dest_id = apicid, + }; - lapic_irq.delivery_mode =APIC_DM_REMRD ;kvm_irq_delivery_to_apic (kvm,NULL , &lapic_irq,NULL ); }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 143/231 net/tls: Check for errors in tls_device_init
In reply to Greg Kroah-HartmanFrom: Tariq Toukan <[email protected]>
[ Upstream commit 3d8c51b25a235e283e37750943bbf356ef187230 ]
Add missing error checks in tls_device_init.
--- include/net/tls.h | 4 ++-- net/tls/tls_device.c | 4 ++-- net/tls/tls_main.c | 7 ++++++- 3 files changed, 10 insertions(+), 5 deletions(-)
include/net/tls.h +2 -2
@@ -708,7 +708,7 @@ int tls_sw_fallback_init(struct sock *sk, struct tls_crypto_info *crypto_info);#ifdef CONFIG_TLS_DEVICE -void tls_device_init (void ); +int tls_device_init (void );void tls_device_cleanup (void );void tls_device_sk_destruct (struct sock *sk);int tls_set_device_offload (struct sock *sk,struct tls_context *ctx);
@@ -728,7 +728,7 @@ static inline bool tls_is_sk_rx_device_offloaded(struct sock *sk)return tls_get_ctx (sk)->rx_conf ==TLS_HW ; }#else -static inline void tls_device_init (void ) {} +static inline int tls_device_init (void ) {return 0 ; }static inline void tls_device_cleanup (void ) {} static inline int
net/tls/tls_device.c +2 -2
@@ -1394,9 +1394,9 @@ static struct notifier_block tls_dev_notifier = { .notifier_call = tls_dev_event, }; -void __init tls_device_init (void ) +int __init tls_device_init (void ) { -register_netdevice_notifier (&tls_dev_notifier); +return register_netdevice_notifier (&tls_dev_notifier); } void __exit tls_device_cleanup(void)
net/tls/tls_main.c +6 -1
@@ -993,7 +993,12 @@ static int __init tls_register(void)if (err)return err; -tls_device_init (); + err =tls_device_init (); +if (err) { +unregister_pernet_subsys (&tls_proc_ops); +return err; + } +tcp_register_ulp (&tcp_tls_ulp_ops);return 0 ;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 144/231 mm: sysctl: fix missing numa_stat when !CONFIG_HUGETLB_PAGE
In reply to Greg Kroah-HartmanFrom: Muchun Song <[email protected]>
[ Upstream commit 43b5240ca6b33108998810593248186b1e3ae34a ]
"numa_stat" should not be included in the scope of CONFIG_HUGETLB_PAGE, if CONFIG_HUGETLB_PAGE is not configured even if CONFIG_NUMA is configured, "numa_stat" is missed form /proc. Move it out of CONFIG_HUGETLB_PAGE to fix it.
--- kernel/sysctl.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-)
kernel/sysctl.c +11 -9
@@ -2466,6 +2466,17 @@ static struct ctl_table vm_table[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_TWO_HUNDRED, }, +#ifdef CONFIG_NUMA + { + .procname = "numa_stat", + .data = &sysctl_vm_numa_stat, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = sysctl_vm_numa_stat_handler, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, +#endif #ifdef CONFIG_HUGETLB_PAGE { .procname = "nr_hugepages",
@@ -2482,15 +2493,6 @@ static struct ctl_table vm_table[] = { .mode = 0644, .proc_handler = &hugetlb_mempolicy_sysctl_handler, }, - { - .procname = "numa_stat", - .data = &sysctl_vm_numa_stat, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = sysctl_vm_numa_stat_handler, - .extra1 = SYSCTL_ZERO, - .extra2 = SYSCTL_ONE, - },#endif { .procname = "hugetlb_shm_group",
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 145/231 x86/kvm: Fix SETcc emulation for return thunks
In reply to Greg Kroah-HartmanFrom: Peter Zijlstra <[email protected]>
[ Upstream commit af2e140f34208a5dfb6b7a8ad2d56bda88f0524d ]
Prepare the SETcc fastop stuff for when RET can be larger still.
The tricky bit here is that the expressions should not only be constant C expressions, but also absolute GAS expressions. This means no ?: and 'true' is ~0.
Also ensure em_setcc() has the same alignment as the actual FOP_SETCC() ops, this ensures there cannot be an alignment hole between em_setcc() and the first op.
Additionally, add a .skip directive to the FOP_SETCC() macro to fill any remaining space with INT3 traps; however the primary purpose of this directive is to generate AS warnings when the remaining space goes negative. Which is a very good indication the alignment magic went side-ways.
--- arch/x86/kvm/emulate.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-)
arch/x86/kvm/emulate.c +15 -13
@@ -325,13 +325,15 @@ static int fastop(struct x86_emulate_ctxt *ctxt, fastop_t fop);#define FOP_RET (name) \ __FOP_RET(#name) -#define FOP_START (op) \ +#define __FOP_START (op, align) \ extern void em_##op(struct fastop *fake); \ asm(".pushsection .text, \"ax\" \n\t" \ ".global em_" #op " \n\t" \ - ".align " __stringify(FASTOP_SIZE) " \n\t" \ + ".align " __stringify(align) " \n\t" \ "em_" #op ":\n\t" +#define FOP_START (op) __FOP_START(op, FASTOP_SIZE) +#define FOP_END \ ".popsection")
@@ -435,16 +437,15 @@ static int fastop(struct x86_emulate_ctxt *ctxt, fastop_t fop);/* * Depending on .config the SETcc functions look like: * -* ENDBR [4 bytes; CONFIG_X86_KERNEL_IBT] -* SETcc %al [3 bytes] -* RET [1 byte] -* INT3 [1 byte; CONFIG_SLS] -* -* Which gives possible sizes 4, 5, 8 or 9. When rounded up to the -* next power-of-two alignment they become 4, 8 or 16 resp. +* ENDBR [4 bytes; CONFIG_X86_KERNEL_IBT] +* SETcc %al [3 bytes] +* RET | JMP __x86_return_thunk [1,5 bytes; CONFIG_RETPOLINE] +* INT3 [1 byte; CONFIG_SLS] */ -#define SETCC_LENGTH (ENDBR_INSN_SIZE + 4 + IS_ENABLED(CONFIG_SLS)) -#define SETCC_ALIGN (4 << IS_ENABLED(CONFIG_SLS) << HAS_KERNEL_IBT) +#define RET_LENGTH (1 + (4 * IS_ENABLED(CONFIG_RETPOLINE)) + \ + IS_ENABLED(CONFIG_SLS)) +#define SETCC_LENGTH (ENDBR_INSN_SIZE + 3 + RET_LENGTH) +#define SETCC_ALIGN (4 << ((SETCC_LENGTH > 4) & 1) << ((SETCC_LENGTH > 8) & 1))static_assert (SETCC_LENGTH <=SETCC_ALIGN );#define FOP_SETCC (op) \
@@ -453,9 +454,10 @@ static_assert(SETCC_LENGTH <= SETCC_ALIGN);#op ": \n\t" \ ASM_ENDBR \ #op " %al \n\t" \ - __FOP_RET(#op) + __FOP_RET(#op) \ + ".skip " __stringify(SETCC_ALIGN) " - (.-" #op "), 0xcc \n\t" -FOP_START(setcc) +__FOP_START(setcc, SETCC_ALIGN) FOP_SETCC(seto) FOP_SETCC(setno) FOP_SETCC(setc)
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 146/231 x86/sev: Avoid using __x86_return_thunk
In reply to Greg Kroah-HartmanFrom: Kim Phillips <[email protected]>
[ Upstream commit 0ee9073000e8791f8b134a8ded31bcc767f7f232 ]
Specifically, it's because __enc_copy() encrypts the kernel after being relocated outside the kernel in sme_encrypt_execute(), and the RET macro's jmp offset isn't amended prior to execution.
--- arch/x86/mm/mem_encrypt_boot.S | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
arch/x86/mm/mem_encrypt_boot.S +6 -2
@@ -65,7 +65,9 @@ SYM_FUNC_START(sme_encrypt_execute) movq %rbp, %rsp /* Restore original stack pointer */ pop %rbp - RET + /* Offset to __x86_return_thunk would be wrong here */ + ret + int3 SYM_FUNC_END(sme_encrypt_execute) SYM_FUNC_START(__enc_copy)
@@ -151,6 +153,8 @@ SYM_FUNC_START(__enc_copy) pop %r12 pop %r15 - RET + /* Offset to __x86_return_thunk would be wrong here */ + ret + int3 .L__enc_copy_end: SYM_FUNC_END(__enc_copy)
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 147/231 x86/bugs: Report AMD retbleed vulnerability
In reply to Greg Kroah-HartmanFrom: Alexandre Chartre <[email protected]>
[ Upstream commit 6b80b59b3555706508008f1f127b5412c89c7fd8 ]
Report that AMD x86 CPUs are vulnerable to the RETBleed (Arbitrary Speculative Code Execution with Return Instructions) attack.
[peterz: add hygon]
[kim: invert parity; fam15h]
--- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/kernel/cpu/bugs.c | 13 +++++++++++++ arch/x86/kernel/cpu/common.c | 19 +++++++++++++++++++ drivers/base/cpu.c | 8 ++++++++ include/linux/cpu.h | 2 ++ 5 files changed, 43 insertions(+)
arch/x86/include/asm/cpufeatures.h +1 -0
@@ -444,5 +444,6 @@#define X86_BUG_ITLB_MULTIHIT X86_BUG(23)/* CPU may incur MCE during certain page attribute changes */ #define X86_BUG_SRBDS X86_BUG(24)/* CPU may leak RNG bits if not mitigated */ #define X86_BUG_MMIO_STALE_DATA X86_BUG(25)/* CPU is affected by Processor MMIO Stale Data vulnerabilities */ +#define X86_BUG_RETBLEED X86_BUG(26)/* CPU is affected by RETBleed */ #endif /* _ASM_X86_CPUFEATURES_H */
arch/x86/kernel/cpu/bugs.c +13 -0
@@ -1987,6 +1987,11 @@ static ssize_t srbds_show_state(char *buf)return sprintf (buf,"%s\n" , srbds_strings[srbds_mitigation]); } +static ssize_t retbleed_show_state (char *buf) +{ +return sprintf (buf,"Vulnerable\n" ); +} +static ssize_t cpu_show_common (struct device *dev,struct device_attribute *attr,char *buf,unsigned int bug) {
@@ -2032,6 +2037,9 @@ static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attrcase X86_BUG_MMIO_STALE_DATA :return mmio_stale_data_show_state (buf); +case X86_BUG_RETBLEED : +return retbleed_show_state (buf); +default :break ; }
@@ -2088,4 +2096,9 @@ ssize_t cpu_show_mmio_stale_data(struct device *dev, struct device_attribute *at {return cpu_show_common (dev, attr, buf,X86_BUG_MMIO_STALE_DATA ); } + +ssize_t cpu_show_retbleed (struct device *dev,struct device_attribute *attr,char *buf) +{ +return cpu_show_common (dev, attr, buf,X86_BUG_RETBLEED ); +}#endif
arch/x86/kernel/cpu/common.c +19 -0
@@ -1231,16 +1231,27 @@ static const __initconst struct x86_cpu_id cpu_vuln_whitelist[] = { {} }; +#define VULNBL (vendor, family, model, blacklist) \ + X86_MATCH_VENDOR_FAM_MODEL(vendor, family, model, blacklist) +#define VULNBL_INTEL_STEPPINGS (model, steppings, issues) \ X86_MATCH_VENDOR_FAM_MODEL_STEPPINGS_FEATURE(INTEL, 6, \ INTEL_FAM6_##model, steppings, \ X86_FEATURE_ANY, issues) +#define VULNBL_AMD (family, blacklist) \ + VULNBL(AMD, family, X86_MODEL_ANY, blacklist) + +#define VULNBL_HYGON (family, blacklist) \ + VULNBL(HYGON, family, X86_MODEL_ANY, blacklist) +#define SRBDS BIT(0)/* CPU is affected by X86_BUG_MMIO_STALE_DATA */ #define MMIO BIT(1)/* CPU is affected by Shared Buffers Data Sampling (SBDS), a variant of X86_BUG_MMIO_STALE_DATA */ #define MMIO_SBDS BIT(2) +/* CPU is affected by RETbleed, speculating where you would not expect it */ +#define RETBLEED BIT(3)static const struct x86_cpu_id cpu_vuln_blacklist[]__initconst = {VULNBL_INTEL_STEPPINGS (IVYBRIDGE ,X86_STEPPING_ANY ,SRBDS ),
@@ -1273,6 +1284,11 @@ static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = {VULNBL_INTEL_STEPPINGS (ATOM_TREMONT ,X86_STEPPINGS (0x1 ,0x1 ),MMIO |MMIO_SBDS ),VULNBL_INTEL_STEPPINGS (ATOM_TREMONT_D ,X86_STEPPING_ANY ,MMIO ),VULNBL_INTEL_STEPPINGS (ATOM_TREMONT_L ,X86_STEPPINGS (0x0 ,0x0 ),MMIO |MMIO_SBDS ), + +VULNBL_AMD (0x15 ,RETBLEED ), +VULNBL_AMD (0x16 ,RETBLEED ), +VULNBL_AMD (0x17 ,RETBLEED ), + VULNBL_HYGON(0x18, RETBLEED), {} };
@@ -1374,6 +1390,9 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) !arch_cap_mmio_immune(ia32_cap))setup_force_cpu_bug (X86_BUG_MMIO_STALE_DATA ); +if (cpu_matches (cpu_vuln_blacklist,RETBLEED )) +setup_force_cpu_bug (X86_BUG_RETBLEED ); +if (cpu_matches (cpu_vuln_whitelist,NO_MELTDOWN ))return ;
drivers/base/cpu.c +8 -0
@@ -570,6 +570,12 @@ ssize_t __weak cpu_show_mmio_stale_data(struct device *dev,return sysfs_emit (buf,"Not affected\n" ); } +ssize_t __weak cpu_show_retbleed (struct device *dev, +struct device_attribute *attr,char *buf) +{ +return sysfs_emit (buf,"Not affected\n" ); +} + static DEVICE_ATTR(meltdown, 0444, cpu_show_meltdown, NULL); static DEVICE_ATTR(spectre_v1, 0444, cpu_show_spectre_v1, NULL); static DEVICE_ATTR(spectre_v2, 0444, cpu_show_spectre_v2, NULL);
@@ -580,6 +586,7 @@ static DEVICE_ATTR(tsx_async_abort, 0444, cpu_show_tsx_async_abort, NULL); static DEVICE_ATTR(itlb_multihit, 0444, cpu_show_itlb_multihit, NULL); static DEVICE_ATTR(srbds, 0444, cpu_show_srbds, NULL); static DEVICE_ATTR(mmio_stale_data, 0444, cpu_show_mmio_stale_data, NULL); +static DEVICE_ATTR(retbleed, 0444, cpu_show_retbleed, NULL);static struct attribute *cpu_root_vulnerabilities_attrs[] = { &dev_attr_meltdown.attr,
@@ -592,6 +599,7 @@ static struct attribute *cpu_root_vulnerabilities_attrs[] = { &dev_attr_itlb_multihit.attr, &dev_attr_srbds.attr, &dev_attr_mmio_stale_data.attr, + &dev_attr_retbleed.attr, NULL };
include/linux/cpu.h +2 -0
@@ -68,6 +68,8 @@ extern ssize_t cpu_show_srbds(struct device *dev, struct device_attribute *attr,extern ssize_t cpu_show_mmio_stale_data (struct device *dev,struct device_attribute *attr,char *buf); +extern ssize_t cpu_show_retbleed (struct device *dev, +struct device_attribute *attr,char *buf); extern __printf(4, 5) struct device *cpu_device_create(struct device *parent, void *drvdata,
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 148/231 objtool: Update Retpoline validation
In reply to Greg Kroah-HartmanFrom: Peter Zijlstra <[email protected]>
[ Upstream commit 9bb2ec608a209018080ca262f771e6a9ff203b6f ]
Update retpoline validation with the new CONFIG_RETPOLINE requirement of not having bare naked RET instructions.
--- arch/x86/include/asm/nospec-branch.h | 6 ++++++ arch/x86/mm/mem_encrypt_boot.S | 2 ++ arch/x86/xen/xen-head.S | 1 + tools/objtool/check.c | 19 +++++++++++++------ 4 files changed, 22 insertions(+), 6 deletions(-)
arch/x86/include/asm/nospec-branch.h +6 -0
@@ -75,6 +75,12 @@ .popsection .endm +/* +* (ab)use RETPOLINE_SAFE on RET to annotate away 'bare' RET instructions +* vs RETBleed validation. +*/ +#define ANNOTATE_UNRET_SAFE ANNOTATE_RETPOLINE_SAFE +/* * JMP_NOSPEC and CALL_NOSPEC macros can be used instead of a simple * indirect jmp/call which may be susceptible to the Spectre variant 2
arch/x86/mm/mem_encrypt_boot.S +2 -0
@@ -66,6 +66,7 @@ SYM_FUNC_START(sme_encrypt_execute) pop %rbp /* Offset to __x86_return_thunk would be wrong here */ + ANNOTATE_UNRET_SAFE ret int3 SYM_FUNC_END(sme_encrypt_execute)
@@ -154,6 +155,7 @@ SYM_FUNC_START(__enc_copy) pop %r15 /* Offset to __x86_return_thunk would be wrong here */ + ANNOTATE_UNRET_SAFE ret int3 .L__enc_copy_end:
arch/x86/xen/xen-head.S +1 -0
@@ -26,6 +26,7 @@ SYM_CODE_START(hypercall_page) .rept (PAGE_SIZE / 32) UNWIND_HINT_FUNC ANNOTATE_NOENDBR + ANNOTATE_UNRET_SAFE ret /* * Xen will write the hypercall page, and sort out ENDBR.
tools/objtool/check.c +13 -6
@@ -2030,8 +2030,9 @@ static int read_retpoline_hints(struct objtool_file *file) }if (insn->type !=INSN_JUMP_DYNAMIC && - insn->type !=INSN_CALL_DYNAMIC ) { -WARN_FUNC ("retpoline_safe hint not an indirect jump/call" , + insn->type !=INSN_CALL_DYNAMIC && + insn->type !=INSN_RETURN ) { +WARN_FUNC ("retpoline_safe hint not an indirect jump/call/ret" , insn->sec, insn->offset);return -1 ; }
@@ -3561,7 +3562,8 @@ static int validate_retpoline(struct objtool_file *file)for_each_insn (file, insn) {if (insn->type !=INSN_JUMP_DYNAMIC && - insn->type !=INSN_CALL_DYNAMIC ) + insn->type !=INSN_CALL_DYNAMIC && + insn->type !=INSN_RETURN )continue ;if (insn->retpoline_safe)
@@ -3576,9 +3578,14 @@ static int validate_retpoline(struct objtool_file *file)if (!strcmp (insn->sec->name,".init.text" ) && !module)continue ; -WARN_FUNC ("indirect %s found in RETPOLINE build" , - insn->sec, insn->offset, - insn->type ==INSN_JUMP_DYNAMIC ?"jump" :"call" ); +if (insn->type ==INSN_RETURN ) { +WARN_FUNC ("'naked' return found in RETPOLINE build" , + insn->sec, insn->offset); + }else { +WARN_FUNC ("indirect %s found in RETPOLINE build" , + insn->sec, insn->offset, + insn->type ==INSN_JUMP_DYNAMIC ?"jump" :"call" ); + } warnings++; }
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 149/231 x86/xen: Rename SYS* entry points
In reply to Greg Kroah-HartmanFrom: Peter Zijlstra <[email protected]>
[ Upstream commit b75b7f8ef1148be1b9321ffc2f6c19238904b438 ]
Native SYS{CALL,ENTER} entry points are called entry_SYS{CALL,ENTER}_{64,compat}, make sure the Xen versions are named consistently.
--- arch/x86/xen/setup.c | 6 +++--- arch/x86/xen/xen-asm.S | 20 ++++++++++---------- arch/x86/xen/xen-ops.h | 6 +++--- 3 files changed, 16 insertions(+), 16 deletions(-)
arch/x86/xen/setup.c +3 -3
@@ -918,7 +918,7 @@ void xen_enable_sysenter(void)if (!boot_cpu_has (sysenter_feature))return ; - ret =register_callback (CALLBACKTYPE_sysenter, xen_sysenter_target); + ret =register_callback (CALLBACKTYPE_sysenter, xen_entry_SYSENTER_compat);if (ret !=0 )setup_clear_cpu_cap (sysenter_feature); }
@@ -927,7 +927,7 @@ void xen_enable_syscall(void) {int ret; - ret =register_callback (CALLBACKTYPE_syscall, xen_syscall_target); + ret =register_callback (CALLBACKTYPE_syscall, xen_entry_SYSCALL_64);if (ret !=0 ) {printk (KERN_ERR "Failed to set syscall callback: %d\n" , ret);/* Pretty fatal; 64-bit userspace has no other
@@ -936,7 +936,7 @@ void xen_enable_syscall(void)if (boot_cpu_has (X86_FEATURE_SYSCALL32 )) { ret =register_callback (CALLBACKTYPE_syscall32, - xen_syscall32_target); + xen_entry_SYSCALL_compat);if (ret !=0 )setup_clear_cpu_cap (X86_FEATURE_SYSCALL32 ); }
arch/x86/xen/xen-asm.S +10 -10
@@ -234,7 +234,7 @@ SYM_CODE_END(xenpv_restore_regs_and_return_to_usermode) */ /* Normal 64-bit system call target */ -SYM_CODE_START(xen_syscall_target) +SYM_CODE_START(xen_entry_SYSCALL_64) UNWIND_HINT_EMPTY ENDBR popq %rcx
@@ -249,12 +249,12 @@ SYM_CODE_START(xen_syscall_target) movq $__USER_CS, 1*8(%rsp) jmp entry_SYSCALL_64_after_hwframe -SYM_CODE_END(xen_syscall_target) +SYM_CODE_END(xen_entry_SYSCALL_64) #ifdef CONFIG_IA32_EMULATION /* 32-bit compat syscall target */ -SYM_CODE_START(xen_syscall32_target) +SYM_CODE_START(xen_entry_SYSCALL_compat) UNWIND_HINT_EMPTY ENDBR popq %rcx
@@ -269,10 +269,10 @@ SYM_CODE_START(xen_syscall32_target) movq $__USER32_CS, 1*8(%rsp) jmp entry_SYSCALL_compat_after_hwframe -SYM_CODE_END(xen_syscall32_target) +SYM_CODE_END(xen_entry_SYSCALL_compat) /* 32-bit compat sysenter target */ -SYM_CODE_START(xen_sysenter_target) +SYM_CODE_START(xen_entry_SYSENTER_compat) UNWIND_HINT_EMPTY ENDBR /*
@@ -291,19 +291,19 @@ SYM_CODE_START(xen_sysenter_target) movq $__USER32_CS, 1*8(%rsp) jmp entry_SYSENTER_compat_after_hwframe -SYM_CODE_END(xen_sysenter_target) +SYM_CODE_END(xen_entry_SYSENTER_compat) #else /* !CONFIG_IA32_EMULATION */ -SYM_CODE_START(xen_syscall32_target) -SYM_CODE_START(xen_sysenter_target) +SYM_CODE_START(xen_entry_SYSCALL_compat) +SYM_CODE_START(xen_entry_SYSENTER_compat) UNWIND_HINT_EMPTY ENDBR lea 16(%rsp), %rsp /* strip %rcx, %r11 */ mov $-ENOSYS, %rax pushq $0 jmp hypercall_iret -SYM_CODE_END(xen_sysenter_target) -SYM_CODE_END(xen_syscall32_target) +SYM_CODE_END(xen_entry_SYSENTER_compat) +SYM_CODE_END(xen_entry_SYSCALL_compat) #endif /* CONFIG_IA32_EMULATION */
arch/x86/xen/xen-ops.h +3 -3
@@ -10,10 +10,10 @@/* These are code, but not functions. Defined in entry.S */ extern const char xen_failsafe_callback[]; -void xen_sysenter_target (void ); +void xen_entry_SYSENTER_compat (void );#ifdef CONFIG_X86_64 -void xen_syscall_target (void ); -void xen_syscall32_target (void ); +void xen_entry_SYSCALL_64 (void ); +void xen_entry_SYSCALL_compat (void );#endif extern void *xen_initial_gdt;
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 150/231 x86/cpu/amd: Add Spectral Chicken
In reply to Greg Kroah-HartmanFrom: Peter Zijlstra <[email protected]>
[ Upstream commit d7caac991feeef1b871ee6988fd2c9725df09039 ]
Zen2 uarchs have an undocumented, unnamed, MSR that contains a chicken bit for some speculation behaviour. It needs setting.
Note: very belatedly AMD released naming; it's now officially called
MSR_AMD64_DE_CFG2 and MSR_AMD64_DE_CFG2_SUPPRESS_NOBR_PRED_BIT but shall remain the SPECTRAL CHICKEN.
--- arch/x86/include/asm/msr-index.h | 3 +++ arch/x86/kernel/cpu/amd.c | 23 ++++++++++++++++++++++- arch/x86/kernel/cpu/cpu.h | 2 ++ arch/x86/kernel/cpu/hygon.c | 6 ++++++ 4 files changed, 33 insertions(+), 1 deletion(-)
arch/x86/include/asm/msr-index.h +3 -0
@@ -552,6 +552,9 @@/* Fam 17h MSRs */ #define MSR_F17H_IRPERF 0xc00000e9 +#define MSR_ZEN2_SPECTRAL_CHICKEN 0xc00110e3 +#define MSR_ZEN2_SPECTRAL_CHICKEN_BIT BIT_ULL(1) +/* Fam 16h MSRs */ #define MSR_F16H_L2I_PERF_CTL 0xc0010230#define MSR_F16H_L2I_PERF_CTR 0xc0010231
arch/x86/kernel/cpu/amd.c +22 -1
@@ -862,6 +862,26 @@ static void init_amd_bd(struct cpuinfo_x86 *c)clear_rdrand_cpuid_bit (c); } +void init_spectral_chicken (struct cpuinfo_x86 *c) +{ +u64 value; + +/* +* On Zen2 we offer this chicken (bit) on the altar of Speculation. +* +* This suppresses speculation from the middle of a basic block, i.e. it +* suppresses non-branch predictions. +* +* We use STIBP as a heuristic to filter out Zen2 from the rest of F17H +*/ +if (!cpu_has (c,X86_FEATURE_HYPERVISOR ) &&cpu_has (c,X86_FEATURE_AMD_STIBP )) { +if (!rdmsrl_safe (MSR_ZEN2_SPECTRAL_CHICKEN , &value)) { + value |=MSR_ZEN2_SPECTRAL_CHICKEN_BIT ; +wrmsrl_safe (MSR_ZEN2_SPECTRAL_CHICKEN , value); + } + } +} +static void init_amd_zn (struct cpuinfo_x86 *c) {set_cpu_cap (c,X86_FEATURE_ZEN );
@@ -907,7 +927,8 @@ static void init_amd(struct cpuinfo_x86 *c)case 0x12 :init_amd_ln (c);break ;case 0x15 :init_amd_bd (c);break ;case 0x16 :init_amd_jg (c);break ; -case 0x17 : fallthrough; +case 0x17 :init_spectral_chicken (c); + fallthrough;case 0x19 :init_amd_zn (c);break ; }
arch/x86/kernel/cpu/cpu.h +2 -0
@@ -61,6 +61,8 @@ static inline void tsx_init(void) { }static inline void tsx_ap_init (void ) { }#endif /* CONFIG_CPU_SUP_INTEL */ +extern void init_spectral_chicken (struct cpuinfo_x86 *c); +extern void get_cpu_cap (struct cpuinfo_x86 *c);extern void get_cpu_address_sizes (struct cpuinfo_x86 *c);extern void cpu_detect_cache_sizes (struct cpuinfo_x86 *c);
arch/x86/kernel/cpu/hygon.c +6 -0
@@ -302,6 +302,12 @@ static void init_hygon(struct cpuinfo_x86 *c)/* get apicid instead of initial apic id from cpuid */ c->apicid =hard_smp_processor_id (); +/* +* XXX someone from Hygon needs to confirm this DTRT +* +init_spectral_chicken(c); +*/ +set_cpu_cap (c,X86_FEATURE_ZEN );set_cpu_cap (c,X86_FEATURE_CPB );
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 151/231 ARM: 9211/1: domain: drop modify_domain()
In reply to Greg Kroah-HartmanFrom: Linus Walleij <[email protected]>
[ Upstream commit cc45b836388f0ccc6831288a08f77a33845f10b0 ]
This function/macro isn't used anywhere in the kernel. The only user was set_fs() and was deleted in the set_fs() removal patch set.
--- arch/arm/include/asm/domain.h | 13 ------------- 1 file changed, 13 deletions(-)
arch/arm/include/asm/domain.h +0 -13
@@ -112,19 +112,6 @@ static __always_inline void set_domain(unsigned int val) }#endif -#ifdef CONFIG_CPU_USE_DOMAINS -#define modify_domain (dom,type) \ - do { \ - unsigned int domain = get_domain(); \ - domain &= ~domain_mask(dom); \ - domain = domain | domain_val(dom, type); \ - set_domain(domain); \ - } while (0) - -#else -static inline void modify_domain (unsigned dom,unsigned type) { } -#endif -/* * Generate the T (user) versions of the LDR/STR and related * instructions (inline assembly)
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 152/231 ARM: 9212/1: domain: Modify Kconfig help text
In reply to Greg Kroah-HartmanFrom: Linus Walleij <[email protected]>
[ Upstream commit 2bf6204240fddb22cc4940b9e3f40c538390212e ]
After the removal of set_fs() the reference to set_fs() is stale. Alter the helptext to reflect what the config option really does.
--- arch/arm/mm/Kconfig | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
arch/arm/mm/Kconfig +5 -1
@@ -631,7 +631,11 @@ config CPU_USE_DOMAINS bool help This option enables or disables the use of domain switching - via the set_fs() function. + using the DACR (domain access control register) to protect memory + domains from each other. In Linux we use three domains: kernel, user + and IO. The domains are used to protect userspace from kernelspace + and to handle IO-space as a special type of memory by assigning + manager or client roles to running code (such as a process). config CPU_V7M_NUM_IRQ int "Number of external interrupts connected to the NVIC"
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 153/231 ASoC: dt-bindings: Fix description for msm8916
In reply to Greg Kroah-HartmanFrom: Bryan O'Donoghue <[email protected]>
[ Upstream commit 94c65dffd4c4af052b3ea8934fbcb2fa8da276a8 ]
For the existing msm8916 bindings the minimum reg/reg-names is 1 not 2. Similarly the minimum interrupt/interrupt-names is 1 not 2.
--- .../devicetree/bindings/sound/qcom,lpass-cpu.yaml | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
Documentation/devicetree/bindings/sound/qcom,lpass-cpu.yaml +4 -4
@@ -25,12 +25,12 @@ properties: - qcom,sc7280-lpass-cpu reg: - minItems: 2 + minItems: 1 maxItems: 6 description: LPAIF core registers reg-names: - minItems: 2 + minItems: 1 maxItems: 6 clocks:
@@ -42,12 +42,12 @@ properties: maxItems: 7 interrupts: - minItems: 2 + minItems: 1 maxItems: 4 description: LPAIF DMA buffer interrupt interrupt-names: - minItems: 2 + minItems: 1 maxItems: 4 qcom,adsp:
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 154/231 tee: tee_get_drvdata(): fix description of return value
In reply to Greg Kroah-HartmanFrom: Marc Kleine-Budde <[email protected]>
[ Upstream commit e5ce073c8a1e01b215a5eb32ba48f8d17ded3bd5 ]
This patch fixes the description of tee_get_drvdata()'s return value. It actually returns the driver_data pointer supplied to tee_device_alloc() since the TEE subsystem was added to the kernel.
--- drivers/tee/tee_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/tee/tee_core.c +1 -1
@@ -1075,7 +1075,7 @@ EXPORT_SYMBOL_GPL(tee_device_unregister);/** * tee_get_drvdata() - Return driver_data pointer * @teedev: Device containing the driver_data pointer -* @returns the driver_data pointer supplied to tee_register(). +* @returns the driver_data pointer supplied to tee_device_alloc(). */ void *tee_get_drvdata (struct tee_device *teedev) {
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 155/231 tty: extract tty_flip_buffer_commit() from tty_flip_buffer_push()
In reply to Greg Kroah-HartmanFrom: Jiri Slaby <[email protected]>
[ Upstream commit 716b10580283fda66f2b88140e3964f8a7f9da89 ]
We will need this new helper in the next patch.
--- drivers/tty/tty_buffer.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-)
drivers/tty/tty_buffer.c +10 -5
@@ -532,6 +532,15 @@ static void flush_to_ldisc(struct work_struct *work) } +static inline void tty_flip_buffer_commit (struct tty_buffer *tail) +{ +/* +* Paired w/ acquire in flush_to_ldisc(); ensures flush_to_ldisc() sees +* buffer data. +*/ +smp_store_release (&tail->commit, tail->used); +} +/** * tty_flip_buffer_push - push terminal buffers * @port: tty port to push
@@ -546,11 +555,7 @@ void tty_flip_buffer_push(struct tty_port *port) {struct tty_bufhead *buf = &port->buf; -/* -* Paired w/ acquire in flush_to_ldisc(); ensures flush_to_ldisc() sees -* buffer data. -*/ -smp_store_release (&buf->tail->commit, buf->tail->used); +tty_flip_buffer_commit (buf->tail);queue_work (system_unbound_wq, &buf->work); }EXPORT_SYMBOL (tty_flip_buffer_push);
Signature · 1 line
2.35.1
Greg Kroah-Hartman
maintainerPATCH 156/231 tty: use new tty_insert_flip_string_and_push_buffer() in pty_write()
In reply to Greg Kroah-HartmanFrom: Jiri Slaby <[email protected]>
[ Upstream commit a501ab75e7624d133a5a3c7ec010687c8b961d23 ]
There is a race in pty_write(). pty_write() can be called in parallel with e.g. ioctl(TIOCSTI) or ioctl(TCXONC) which also inserts chars to the buffer. Provided, tty_flip_buffer_push() in pty_write() is called outside the lock, it can commit inconsistent tail. This can lead to out of bounds writes and other issues. See the Link below.
To fix this, we have to introduce a new helper called tty_insert_flip_string_and_push_buffer(). It does both tty_insert_flip_string() and tty_flip_buffer_commit() under the port lock. It also calls queue_work(), but outside the lock. See 71a174b39f10 (pty: do tty_flip_buffer_push without port->lock in pty_write) for the reasons.
Keep the helper internal-only (in drivers' tty.h). It is not intended to be used widely.
--- drivers/tty/pty.c | 14 ++------------ drivers/tty/tty.h | 3 +++ drivers/tty/tty_buffer.c | 31 +++++++++++++++++++++++++++++++ 3 files changed, 36 insertions(+), 12 deletions(-)
drivers/tty/pty.c +2 -12
@@ -111,21 +111,11 @@ static void pty_unthrottle(struct tty_struct *tty)static int pty_write (struct tty_struct *tty,const unsigned char *buf,int c) {struct tty_struct *to = tty->link; -unsigned long flags; -if (tty->flow.stopped) +if (tty->flow.stopped || !c)return 0 ; -if (c >0 ) { -spin_lock_irqsave (&to->port->lock, flags); -/* Stuff the data into the input queue of the other end */ - c =tty_insert_flip_string (to->port, buf, c); -spin_unlock_irqrestore (&to->port->lock, flags); -/* And shovel */ -if (c) -tty_flip_buffer_push (to->port); - } -return c; +return tty_insert_flip_string_and_push_buffer (to->port, buf, c); }/**
drivers/tty/tty.h +3 -0
@@ -111,4 +111,7 @@ static inline void tty_audit_tiocsti(struct tty_struct *tty, char ch)ssize_t redirected_tty_write (struct kiocb *,struct iov_iter *); +int tty_insert_flip_string_and_push_buffer (struct tty_port *port, +const unsigned char *chars,size_t cnt); +#endif
drivers/tty/tty_buffer.c +31 -0
@@ -560,6 +560,37 @@ void tty_flip_buffer_push(struct tty_port *port) }EXPORT_SYMBOL (tty_flip_buffer_push); +/** +* tty_insert_flip_string_and_push_buffer - add characters to the tty buffer and +* push +* @port: tty port +* @chars: characters +* @size: size +* +* The function combines tty_insert_flip_string() and tty_flip_buffer_push() +* with the exception of properly holding the @port->lock. +* +* To be used only internally (by pty currently). +* +* Returns: the number added. +*/ +int tty_insert_flip_string_and_push_buffer (struct tty_port *port, +const unsigned char *chars,size_t size) +{ +struct tty_bufhead *buf = &port->buf; +unsigned long flags; + +spin_lock_irqsave (&port->lock, flags); + size =tty_insert_flip_string (port, chars, size); +if (size) +tty_flip_buffer_commit (buf->tail); +spin_unlock_irqrestore (&port->lock, flags); + +queue_work (system_unbound_wq, &buf->work); + +return size; +} +/** * tty_buffer_init - prepare a tty buffer structure * @port: tty port to initialise
Signature · 1 line
2.35.1