Skip to contentSkip to conversation

6.8.9-rc1 review

PATCH In review Greg Kroah-Hartman 12 participants 240 messages Apr 30, 2024 – May 2, 2024 Page 1 of 2

Subsystems (11)
lore
Share & tools
lore

Greg Kroah-Hartman

author maintainer
#0 lore raw

cover 6.8.9-rc1 review

This is the start of the stable review cycle for the 6.8.9 release. There are 228 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, 02 May 2024 10:30:27 +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/v6.x/stable-review/patch-6.8.9-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-6.8.y
and the diffstat can be found below.

thanks,

greg k-h

-------------
Pseudo-Shortlog of commits:

Greg Kroah-Hartman <[email protected]>

    Linux 6.8.9-rc1

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Fix UAF on hci_abort_conn_sync

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Fix UAF on create_le_conn_complete

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Fix UAF in hci_acl_create_conn_sync

Matthew Wilcox (Oracle) <[email protected]>
bounds: Use the right number of bits for power-of-two CONFIG_NR_CPUS

Johan Hovold <[email protected]>
phy: qcom: qmp-combo: fix VCO div offset on v5_5nm and v6

Wolfram Sang <[email protected]>
i2c: smbus: fix NULL function pointer dereference

Andrew Jones <[email protected]>
RISC-V: selftests: cbo: Ensure asm operands match constraints, take 2

Clément Léger <[email protected]>
riscv: hwprobe: fix invalid sign extension for RISCV_HWPROBE_EXT_ZVFHMIN

Xuewen Yan <[email protected]>
sched/eevdf: Prevent vlag from going out of bounds in reweight_eevdf()

Tianchen Ding <[email protected]>
sched/eevdf: Fix miscalculation in reweight_entity() when se is not curr

Tianchen Ding <[email protected]>

    sched/eevdf: Always update V if se->on_rq when reweighting

Rob Herring <[email protected]>
dt-bindings: eeprom: at24: Fix ST M24C64-D compatible schema

Hans de Goede <[email protected]>
phy: ti: tusb1210: Resolve charger-det crash if charger psy is unregistered

Samuel Holland <[email protected]>
riscv: Fix loading 64-bit NOMMU kernels past the start of RAM

Samuel Holland <[email protected]>
riscv: Fix TASK_SIZE on 64-bit NOMMU

Fenghua Yu <[email protected]>
dmaengine: idxd: Fix oops during rmmod on single-CPU platforms

Sean Anderson <[email protected]>
dma: xilinx_dpdma: Fix locking

Rex Zhang <[email protected]>
dmaengine: idxd: Convert spinlock to mutex to lock evl workqueue

Gabor Juhos <[email protected]>
phy: qcom: m31: match requested regulator name with dt schema

Sebastian Reichel <[email protected]>
phy: rockchip: naneng-combphy: Fix mux on rk3588

Sebastian Reichel <[email protected]>
phy: rockchip-snps-pcie3: fix clearing PHP_GRF_PCIESEL_CON bits

Michal Tomek <[email protected]>
phy: rockchip-snps-pcie3: fix bifurcation on rk3588

Marcel Ziswiler <[email protected]>
phy: freescale: imx8m-pcie: fix pcie link-up instability

Mikhail Kobuk <[email protected]>
phy: marvell: a3700-comphy: Fix hardcoded array size

Mikhail Kobuk <[email protected]>
phy: marvell: a3700-comphy: Fix out of bounds read

Vijendar Mukunda <[email protected]>
soundwire: amd: fix for wake interrupt handling for clockstop mode

Andy Shevchenko <[email protected]>
idma64: Don't try to serve interrupts when device is powered off

Akhil R <[email protected]>
dmaengine: tegra186: Fix residual calculation

Arnd Bergmann <[email protected]>
dmaengine: owl: fix register access functions

Johannes Weiner <[email protected]>
mm: zswap: fix shrinker NULL crash with cgroup_disable=memory

Matthew Wilcox (Oracle) <[email protected]>
mm: turn folio_test_hugetlb into a PageType

Stephen Boyd <[email protected]>
phy: qcom: qmp-combo: Fix VCO div offset on v3

Stephen Boyd <[email protected]>
phy: qcom: qmp-combo: Fix register base for QSERDES_DP_PHY_MODE

Maximilian Luz <[email protected]>
firmware: qcom: uefisecapp: Fix memory related IO errors and crashes

Arnd Bergmann <[email protected]>
mtd: diskonchip: work around ubsan link failure

Christian Marangi <[email protected]>
mtd: limit OTP NVMEM cell parse to non-NAND devices

Yick Xie <[email protected]>
udp: preserve the connected status if only UDP cmsg

Nam Cao <[email protected]>
fbdev: fix incorrect address computation in deferred IO

Andrey Ryabinin <[email protected]>
stackdepot: respect __GFP_NOLOCKDEP allocation flag

Rahul Rameshbabu <[email protected]>
macsec: Detect if Rx skb is macsec-related for offloading devices that update md_dst

Rahul Rameshbabu <[email protected]>
macsec: Enable devices to advertise whether they update sk_buff md_dst during offloads

Peter Münster <[email protected]>
net: b44: set pause params only when interface is up

Rahul Rameshbabu <[email protected]>
ethernet: Add helper for assigning packet type when dest address does not match device address

Rahul Rameshbabu <[email protected]>
net/mlx5e: Advertise mlx5 ethernet driver updates sk_buff md_dst for MACsec

Vanshidhar Konda <[email protected]>
ACPI: CPPC: Fix access width used for PCC registers

Jarred White <[email protected]>
ACPI: CPPC: Fix bit_offset shift in MASK_VAL() macro

Jarred White <[email protected]>
ACPI: CPPC: Use access_width over bit_width for system memory accesses

Guanrui Huang <[email protected]>
irqchip/gic-v3-its: Prevent double free on error

Felix Kuehling <[email protected]>
drm/amdkfd: Fix eviction fence handling

Felix Kuehling <[email protected]>
drm/amdkfd: Fix rescheduling of restore worker

Mukul Joshi <[email protected]>
drm/amdgpu: Fix leak when GPU memory allocation fails

Lang Yu <[email protected]>
drm/amdgpu/umsch: don't execute umsch test when GPU is in reset/suspend

Ma Jun <[email protected]>
drm/amdgpu/pm: Remove gpu_od if it's an empty directory

Lucas Stach <[email protected]>
drm/atomic-helper: fix parameter order in drm_format_conv_state_copy() call

Lijo Lazar <[email protected]>
drm/amdgpu: Assign correct bits for SDMA HDP flush

Alex Deucher <[email protected]>
drm/amdgpu/sdma5.2: use legacy HDP flush for SDMA2/3

Louis Chauvet <[email protected]>
dmaengine: xilinx: xdma: Fix synchronization issue

Miquel Raynal <[email protected]>
dmaengine: xilinx: xdma: Fix wrong offsets in the buffers addresses in dma descriptor

Vinod Koul <[email protected]>
dmaengine: Revert "dmaengine: pl330: issue_pending waits until WFP state"

Iskander Amara <[email protected]>
arm64: dts: rockchip: enable internal pull-up for Q7_THRM# on RK3399 Puma

Manivannan Sadhasivam <[email protected]>
arm64: dts: qcom: sm8450: Fix the msi-map entries

Johan Hovold <[email protected]>

    arm64: dts: qcom: sc8280xp: add missing PCIe minimum OPP

Jiantao Shan <[email protected]>
LoongArch: Fix access error when read fault on a write-only VMA

Huacai Chen <[email protected]>
LoongArch: Fix callchain parse error with kernel tracepoint events

Daniel Okazaki <[email protected]>
eeprom: at24: fix memory corruption race condition

Sean Christopherson <[email protected]>
cpu: Re-enable CPU mitigations by default for !X86 architectures

Kirill A. Shutemov <[email protected]>
x86/tdx: Preserve shared bit on mprotect()

Johannes Thumshirn <[email protected]>
btrfs: fix information leak in btrfs_ioctl_logical_to_ino()

Naohiro Aota <[email protected]>
btrfs: scrub: run relocation repair when/only needed

Qu Wenruo <[email protected]>
btrfs: fix wrong block_start calculation for btrfs_drop_extent_map_range()

Sweet Tea Dorminy <[email protected]>
btrfs: fallback if compressed IO fails for ENOSPC

Kenny Levinsen <[email protected]>
HID: i2c-hid: Revert to await reset ACK before reading report descriptor

Nam Cao <[email protected]>
HID: i2c-hid: remove I2C_HID_READ_PENDING flag to prevent lock-up

Steve French <[email protected]>
smb3: fix lock ordering potential deadlock in cifs_sync_mid_result

Steve French <[email protected]>
smb3: missing lock when picking channel

Gustavo A. R. Silva <[email protected]>
smb: client: Fix struct_group() usage in __packed structs

Miaohe Lin <[email protected]>
mm/hugetlb: fix DEBUG_LOCKS_WARN_ON(1) when dissolve_free_hugetlb_folio()

Matthew Wilcox (Oracle) <[email protected]>
mm: support page_mapcount() on page_has_type() pages

Matthew Wilcox (Oracle) <[email protected]>
mm: create FOLIO_FLAG_FALSE and FOLIO_TYPE_OPS macros

Maksim Kiselev <[email protected]>
mmc: sdhci-of-dwcmshc: th1520: Increase tuning loop count to 128

Mantas Pucka <[email protected]>
mmc: sdhci-msm: pervent access to suspended controller

Peter Xu <[email protected]>
mm/hugetlb: fix missing hugetlb_lock for resv uncharge

Christian Marangi <[email protected]>
mtd: rawnand: qcom: Fix broken OP_RESET_DEVICE command in qcom_misc_cmd_type_exec()

Johan Hovold <[email protected]>
Bluetooth: qca: fix NULL-deref on non-serdev setup

Johan Hovold <[email protected]>
Bluetooth: qca: fix NULL-deref on non-serdev suspend

WangYuli <[email protected]>

    Bluetooth: btusb: Add Realtek RTL8852BE support ID 0x0bda:0x4853

Nathan Chancellor <[email protected]>
Bluetooth: Fix type of len in {l2cap,sco}_sock_getsockopt_old()

Aswin Unnikrishnan <[email protected]>
rust: remove params from module macro example

Miguel Ojeda <[email protected]>
kbuild: rust: force alloc extern to allow "empty" Rust files

Miguel Ojeda <[email protected]>
kbuild: rust: remove unneeded @rustc_cfg to avoid ICE

Conor Dooley <[email protected]>
rust: make mutually exclusive with CFI_CLANG

Laine Taffin Altman <[email protected]>
rust: init: remove impl Zeroable for Infallible

Alice Ryhl <[email protected]>
rust: don't select CONSTRUCTORS

Wedson Almeida Filho <[email protected]>
rust: kernel: require Send for Module implementations

Wedson Almeida Filho <[email protected]>
rust: phy: implement Send for Registration

David Kaplan <[email protected]>
x86/cpu: Fix check for RDPKRU in __show_regs()

Wenkuan Wang <[email protected]>

    x86/CPU/AMD: Add models 0x10-0x1f to the Zen5 range

Terry Tritton <[email protected]>
selftests/seccomp: Handle EINVAL on unshare(CLONE_NEWPID)

Terry Tritton <[email protected]>
selftests/seccomp: Change the syscall used in KILL_THREAD test

Terry Tritton <[email protected]>
selftests/seccomp: user_notification_addfd check nextfd is available

Christian König <[email protected]>
drm/amdgpu: fix visible VRAM handling during faults

Alex Deucher <[email protected]>
drm/amdgpu: add shared fdinfo stats

Alex Deucher <[email protected]>
drm: add drm_gem_object_is_shared_for_memory_stats() helper

Sean Christopherson <[email protected]>
KVM: x86/pmu: Set enable bits for GP counters in PERF_GLOBAL_CTRL at "RESET"

Sean Christopherson <[email protected]>
KVM: x86/pmu: Zero out PMU metadata on AMD if PMU is disabled

Oliver Neukum <[email protected]>
usb: xhci: correct return value in case of STS_HCE

Mathias Nyman <[email protected]>
xhci: move event processing for one interrupter to a separate function

Alex Deucher <[email protected]>
Revert "drm/amd/display: fix USB-C flag update after enc10 feature init"

George Shen <[email protected]>
drm/amd/display: Check DP Alt mode DPCS state via DMUB

David Howells <[email protected]>
netfs: Fix the pre-flush when appending to a file in writethrough mode

Kuniyuki Iwashima <[email protected]>
af_unix: Suppress false-positive lockdep splat for spin_lock() in __unix_gc().

Sabrina Dubroca <[email protected]>

    tls: fix lockless read of strp->msg_ready in ->poll

Arkadiusz Kubalewski <[email protected]>
dpll: fix dpll_pin_on_pin_register() for multiple parent pins

Jiri Pirko <[email protected]>
dpll: check that pin is registered in __dpll_pin_unregister()

Su Hui <[email protected]>
octeontx2-af: fix the double free in rvu_npc_freemem()

Jason Reeder <[email protected]>
net: ethernet: ti: am65-cpts: Fix PTPv1 message type on TX packets

Jacob Keller <[email protected]>
ice: fix LAG and VF lock dependency in ice_reset_vf()

Sudheer Mogilappagari <[email protected]>
iavf: Fix TC config comparison with existing adapter TC config

Erwan Velu <[email protected]>
i40e: Report MFS in decimal base instead of hex

Sindhu Devale <[email protected]>
i40e: Do not use WQ_MEM_RECLAIM flag for workqueue

Dan Carpenter <[email protected]>
net: ti: icssg-prueth: Fix signedness bug in prueth_init_rx_chns()

MD Danish Anwar <[email protected]>
net: phy: dp83869: Fix MII mode failure

Pablo Neira Ayuso <[email protected]>
netfilter: nf_tables: honor table dormant flag from netdev release event path

Michael Heimpold <[email protected]>
ARM: dts: imx6ull-tarragon: fix USB over-current polarity

Jakub Kicinski <[email protected]>
eth: bnxt: fix counting packets discarded due to OOM and netpoll

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix memory leak when canceling rehash work

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix incorrect list API usage

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix warning during rehash

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix memory leak during rehash

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Rate limit error message

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix possible use-after-free during rehash

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix possible use-after-free during activity update

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix race during rehash delayed work

Ido Schimmel <[email protected]>
mlxsw: spectrum_acl_tcam: Fix race in region ID allocation

Amit Cohen <[email protected]>
mlxsw: Use refcount_t for reference counting

Hyunwoo Kim <[email protected]>
net: openvswitch: Fix Use-After-Free in ovs_ct_exit

Ismael Luceno <[email protected]>
ipvs: Fix checksumming on GSO of SCTP packets

Bartosz Golaszewski <[email protected]>
Bluetooth: qca: set power_ctrl_enabled on NULL returned by gpiod_get_optional()

Chun-Yi Lee <[email protected]>
Bluetooth: hci_sync: Using hci_cmd_sync_submit when removing Adv Monitor

Sean Wang <[email protected]>
Bluetooth: btusb: mediatek: Fix double free of skb in coredump

Luiz Augusto von Dentz <[email protected]>
Bluetooth: MGMT: Fix failing to MGMT_OP_ADD_UUID/MGMT_OP_REMOVE_UUID

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_event: Fix sending HCI_OP_READ_ENC_KEY_SIZE

Zijun Hu <[email protected]>
Bluetooth: btusb: Fix triggering coredump implementation for QCA

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync

Iulia Tanasescu <[email protected]>
Bluetooth: ISO: Reassemble PA data for bcast sink

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Attempt to dequeue connection attempt

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_sync: Add helper functions to manipulate cmd_sync queue

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_conn: Fix UAF Write in __hci_acl_create_connection_sync

Luiz Augusto von Dentz <[email protected]>
Bluetooth: hci_conn: Always use sk_timeo as conn_timeout

Jonas Dreßler <[email protected]>
Bluetooth: Remove pending ACL connection attempts

Jonas Dreßler <[email protected]>
Bluetooth: hci_conn: Only do ACL connections sequentially

Jonas Dreßler <[email protected]>
Bluetooth: hci_event: Use HCI error defines instead of magic values

Himal Prasad Ghimiray <[email protected]>
drm/xe: call free_gsc_pkt only once on action add failure

Himal Prasad Ghimiray <[email protected]>
drm/xe: Remove sysfs only once on action add failure

Prathamesh Shete <[email protected]>
gpio: tegra186: Fix tegra186_gpio_is_accessible() check

Daniel Golle <[email protected]>
net: phy: mediatek-ge-soc: follow netdev LED trigger semantics

Hyunwoo Kim <[email protected]>
net: gtp: Fix Use-After-Free in gtp_dellink

Hyunwoo Kim <[email protected]>
tcp: Fix Use-After-Free in tcp_ao_connect_init

Eric Dumazet <[email protected]>

    net: usb: ax88179_178a: stop lying about skb->truesize

Eric Dumazet <[email protected]>
ipv4: check for NULL idev in ip_route_use_hint()

Eric Dumazet <[email protected]>
net: fix sk_memory_allocated_{add|sub} vs softirqs

Adam Li <[email protected]>
net: make SK_MEMORY_PCPU_RESERV tunable

Jakub Kicinski <[email protected]>
tools: ynl: don't ignore errors in NLMSG_DONE messages

AngeloGioacchino Del Regno <[email protected]>
soc: mediatek: mtk-svs: Append "-thermal" to thermal zone names

Duoming Zhou <[email protected]>
ax25: Fix netdev refcount issue

David Howells <[email protected]>
netfs: Fix writethrough-mode error handling

Paul Geurts <[email protected]>
NFC: trf7970a: disable all regulators on removal

Matthias Schiffer <[email protected]>
net: dsa: mv88e6xx: fix supported_interfaces setup in mv88e6250_phylink_get_caps()

Dan Williams <[email protected]>
cxl/core: Fix potential payload size confusion in cxl_mem_get_poison()

Michael Chan <[email protected]>
bnxt_en: Fix error recovery for 5760X (P7) chips

Vikas Gupta <[email protected]>
bnxt_en: Fix the PCI-AER routines

Vikas Gupta <[email protected]>
bnxt_en: refactor reset close code

Hangbin Liu <[email protected]>
bridge/br_netlink.c: no need to return void function

Eric Dumazet <[email protected]>
icmp: prevent possible NULL dereferences from icmp_build_probe()

Andrei Simion <[email protected]>
ARM: dts: microchip: at91-sama7g5ek: Replace regulator-suspend-voltage with the valid property

Ido Schimmel <[email protected]>
mlxsw: pci: Fix driver initialization with old firmware

Ido Schimmel <[email protected]>
mlxsw: core_env: Fix driver initialization with old firmware

Ido Schimmel <[email protected]>
mlxsw: core: Unregister EMAD trap using FORWARD action

Justin Chen <[email protected]>
net: bcmasp: fix memory leak when bringing down interface

David Bauer <[email protected]>
vxlan: drop packets from invalid src-address

Duanqiang Wen <[email protected]>
net: libwx: fix alloc msix vectors failed

Johannes Berg <[email protected]>
wifi: mac80211: fix unaligned le16 access

Johannes Berg <[email protected]>
wifi: mac80211: remove link before AP

Johannes Berg <[email protected]>
wifi: mac80211_hwsim: init peer measurement result

Patrik Jakobsson <[email protected]>
drm/gma500: Remove lid code

Miri Korenblit <[email protected]>
wifi: iwlwifi: mvm: return uid from iwl_mvm_build_scan_cmd

Avraham Stern <[email protected]>
wifi: iwlwifi: mvm: remove old PASN station when adding a new one

Felix Fietkau <[email protected]>
wifi: mac80211: split mesh fast tx cache into local/proxied/forwarded

Colin Ian King <[email protected]>
wifi: mac80211: clean up assignments to pointer cache.

Alexey Brodkin <[email protected]>
ARC: [plat-hsdk]: Remove misplaced interrupt-cells property

Andy Shevchenko <[email protected]>
gpio: tangier: Use correct type for the IRQ chip data

Maximilian Luz <[email protected]>

    arm64: dts: qcom: sc8180x: Fix ss_phy_irq for secondary USB controller

Rajendra Nayak <[email protected]>
arm64: dts: qcom: x1e80100: Fix the compatible for cluster idle states

Luca Weiss <[email protected]>
arm64: dts: qcom: Fix type of "wdog" IRQs for remoteprocs

Yu Kuai <[email protected]>
block: fix module reference leakage from bdev_open_by_dev error path

Jose Ignacio Tornos Martinez <[email protected]>
arm64: dts: rockchip: regulator for sd needs to be always on for BPI-R2Pro

Muhammed Efe Cetin <[email protected]>
arm64: dts: rockchip: mark system power controller and fix typo on orangepi-5-plus

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt2712: fix validation errors

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7986: prefix BPI-R3 cooling maps with "map-"

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7986: drop invalid thermal block clock

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7986: reorder nodes

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7986: drop "#reset-cells" from Ethernet controller

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7986: drop invalid properties from ethsys

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7986: reorder properties

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7622: drop "reset-names" from thermal block

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7622: fix ethernet controller "compatible"

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7622: fix IR nodename

Rafał Miłecki <[email protected]>
arm64: dts: mediatek: mt7622: fix clock controllers

Pin-yen Lin <[email protected]>
arm64: dts: mediatek: mt8183-kukui: Use default min voltage for MT6358

Pin-yen Lin <[email protected]>
arm64: dts: mediatek: mt8195-cherry: Update min voltage constraint for MT6315

Pin-yen Lin <[email protected]>
arm64: dts: mediatek: mt8192-asurada: Update min voltage constraint for MT6315

Nícolas F. R. A. Prado <[email protected]>
arm64: dts: mediatek: cherry: Describe CPU supplies

Nícolas F. R. A. Prado <[email protected]>
arm64: dts: mediatek: mt8195: Add missing gce-client-reg to mutex1

Nícolas F. R. A. Prado <[email protected]>
arm64: dts: mediatek: mt8195: Add missing gce-client-reg to mutex

Nícolas F. R. A. Prado <[email protected]>
arm64: dts: mediatek: mt8195: Add missing gce-client-reg to vpp/vdosys

Nícolas F. R. A. Prado <[email protected]>
arm64: dts: mediatek: mt8192: Add missing gce-client-reg to mutex

Ikjoon Jang <[email protected]>
arm64: dts: mediatek: mt8183: Add power-domains properity to mfgcfg

Dragan Simic <[email protected]>
arm64: dts: rockchip: Remove unsupported node from the Pinebook Pro dts

Andy Yan <[email protected]>
arm64: dts: rockchip: Fix the i2c address of es8316 on Cool Pi CM5

Quentin Schulz <[email protected]>
arm64: dts: rockchip: enable internal pull-up on PCIE_WAKE# for RK3399 Puma

Iskander Amara <[email protected]>
arm64: dts: rockchip: fix alphabetical ordering RK3399 puma

Quentin Schulz <[email protected]>
arm64: dts: rockchip: enable internal pull-up on Q7_USB_ID for RK3399 Puma

Arınç ÜNAL <[email protected]>

    arm64: dts: rockchip: set PHY address of MT7531 switch to 0x1f

Yaraslau Furman <[email protected]>
HID: logitech-dj: allow mice to use all types of reports

Zhang Lixu <[email protected]>

    HID: intel-ish-hid: ipc: Fix dev_err usage with uninitialized dev->devc

Takayuki Nagata <[email protected]>
cifs: reinstate original behavior again for forceuid/forcegid

Paulo Alcantara <[email protected]>
smb: client: fix rename(2) regression against samba

David Howells <[email protected]>
cifs: Fix reacquisition of volume cookie on still-live connection

-------------

Diffstat:

 Documentation/admin-guide/sysctl/net.rst           |   5 +
 Documentation/devicetree/bindings/eeprom/at24.yaml |   5 +-
 Makefile                                           |   4 +-
 arch/Kconfig                                       |   8 +
 arch/arc/boot/dts/hsdk.dts                         |   1 -
 arch/arm/boot/dts/microchip/at91-sama7g5ek.dts     |   8 +-
 .../boot/dts/nxp/imx/imx6ull-tarragon-common.dtsi  |   1 +
 arch/arm64/boot/dts/mediatek/mt2712-evb.dts        |   8 +-
 arch/arm64/boot/dts/mediatek/mt2712e.dtsi          |   3 +-
 arch/arm64/boot/dts/mediatek/mt7622.dtsi           |  34 ++-
 .../boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dts  |   6 +-
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi          | 215 ++++++++--------
 arch/arm64/boot/dts/mediatek/mt8183-kukui.dtsi     |   1 -
 arch/arm64/boot/dts/mediatek/mt8183.dtsi           |   1 +
 arch/arm64/boot/dts/mediatek/mt8192-asurada.dtsi   |   6 +-
 arch/arm64/boot/dts/mediatek/mt8192.dtsi           |   1 +
 arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi    |  36 ++-
 arch/arm64/boot/dts/mediatek/mt8195.dtsi           |   5 +
 arch/arm64/boot/dts/qcom/sc7280.dtsi               |   4 +-
 arch/arm64/boot/dts/qcom/sc8180x.dtsi              |   2 +-
 arch/arm64/boot/dts/qcom/sc8280xp.dtsi             |  11 +-
 arch/arm64/boot/dts/qcom/sm6350.dtsi               |   4 +-
 arch/arm64/boot/dts/qcom/sm6375.dtsi               |   2 +-
 arch/arm64/boot/dts/qcom/sm8250.dtsi               |   6 +-
 arch/arm64/boot/dts/qcom/sm8450.dtsi               |  16 +-
 arch/arm64/boot/dts/qcom/x1e80100.dtsi             |   4 +-
 .../boot/dts/rockchip/rk3399-pinebook-pro.dts      |   1 -
 arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi      |  31 ++-
 arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts |   6 +-
 .../arm64/boot/dts/rockchip/rk3588-coolpi-cm5.dtsi |   4 +-
 .../boot/dts/rockchip/rk3588-orangepi-5-plus.dts   |   3 +-
 arch/loongarch/include/asm/perf_event.h            |   8 +
 arch/loongarch/mm/fault.c                          |   4 +-
 arch/riscv/include/asm/page.h                      |   2 +-
 arch/riscv/include/asm/pgtable.h                   |   2 +-
 arch/riscv/include/uapi/asm/hwprobe.h              |   2 +-
 arch/riscv/mm/init.c                               |   2 +-
 arch/x86/Kconfig                                   |  11 +-
 arch/x86/include/asm/coco.h                        |   1 +
 arch/x86/include/asm/pgtable_types.h               |   3 +-
 arch/x86/kernel/cpu/amd.c                          |   3 +-
 arch/x86/kernel/process_64.c                       |   2 +-
 arch/x86/kvm/pmu.c                                 |  30 ++-
 arch/x86/kvm/vmx/pmu_intel.c                       |  16 +-
 block/bdev.c                                       |   2 +-
 drivers/acpi/cppc_acpi.c                           |  72 ++++--
 drivers/bluetooth/btmtk.c                          |   7 +-
 drivers/bluetooth/btusb.c                          |  11 +-
 drivers/bluetooth/hci_qca.c                        |  27 +-
 drivers/cxl/core/mbox.c                            |  38 ++-
 drivers/dma/idma64.c                               |   4 +
 drivers/dma/idxd/cdev.c                            |   5 +-
 drivers/dma/idxd/debugfs.c                         |   4 +-
 drivers/dma/idxd/device.c                          |   8 +-
 drivers/dma/idxd/idxd.h                            |   2 +-
 drivers/dma/idxd/init.c                            |   2 +-
 drivers/dma/idxd/irq.c                             |   4 +-
 drivers/dma/idxd/perfmon.c                         |   9 +-
 drivers/dma/owl-dma.c                              |   4 +-
 drivers/dma/pl330.c                                |   3 -
 drivers/dma/tegra186-gpc-dma.c                     |   3 +
 drivers/dma/xilinx/xdma-regs.h                     |   3 +
 drivers/dma/xilinx/xdma.c                          |  28 ++-
 drivers/dma/xilinx/xilinx_dpdma.c                  |  13 +-
 drivers/dpll/dpll_core.c                           |  61 +++--
 drivers/firmware/qcom/qcom_qseecom_uefisecapp.c    | 137 +++++++----
 drivers/firmware/qcom/qcom_scm.c                   |  37 +--
 drivers/gpio/gpio-tangier.c                        |   9 +-
 drivers/gpio/gpio-tegra186.c                       |  20 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c   |   1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c             |   2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c         |   4 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c         |  33 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h         |  28 +--
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c            |  73 +++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h            |   3 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c       |   3 +
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c           |   3 +-
 drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c             |  24 +-
 drivers/gpu/drm/amd/amdkfd/kfd_process.c           |  15 +-
 .../amd/display/dc/dcn32/dcn32_dio_link_encoder.c  |  85 +++++--
 .../amd/display/dc/dcn32/dcn32_dio_link_encoder.h  |   5 +
 .../amd/display/dc/dcn35/dcn35_dio_link_encoder.c  |   4 +-
 drivers/gpu/drm/amd/pm/amdgpu_pm.c                 |   7 +
 drivers/gpu/drm/drm_gem_atomic_helper.c            |   4 +-
 drivers/gpu/drm/gma500/Makefile                    |   1 -
 drivers/gpu/drm/gma500/psb_device.c                |   5 +-
 drivers/gpu/drm/gma500/psb_drv.h                   |   9 -
 drivers/gpu/drm/gma500/psb_lid.c                   |  80 ------
 drivers/gpu/drm/xe/xe_gt.c                         |   4 +-
 drivers/gpu/drm/xe/xe_gt_ccs_mode.c                |  19 +-
 drivers/gpu/drm/xe/xe_gt_ccs_mode.h                |   2 +-
 drivers/gpu/drm/xe/xe_huc.c                        |   9 +-
 drivers/hid/hid-logitech-dj.c                      |   4 +-
 drivers/hid/i2c-hid/i2c-hid-core.c                 |  38 +--
 drivers/hid/intel-ish-hid/ipc/ipc.c                |   2 +-
 drivers/i2c/i2c-core-base.c                        |  12 +-
 drivers/irqchip/irq-gic-v3-its.c                   |   9 +-
 drivers/misc/eeprom/at24.c                         |  18 +-
 drivers/mmc/host/sdhci-msm.c                       |  16 +-
 drivers/mmc/host/sdhci-of-dwcmshc.c                |   1 +
 drivers/mtd/mtdcore.c                              |   2 +-
 drivers/mtd/nand/raw/diskonchip.c                  |   4 +-
 drivers/mtd/nand/raw/qcom_nandc.c                  |   7 +-
 drivers/net/dsa/mv88e6xxx/chip.c                   |  56 ++++-
 drivers/net/dsa/mv88e6xxx/port.h                   |  23 +-
 drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c   |  21 +-
 drivers/net/ethernet/broadcom/b44.c                |  14 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c          |  82 +++---
 drivers/net/ethernet/intel/i40e/i40e_main.c        |   6 +-
 drivers/net/ethernet/intel/iavf/iavf_main.c        |  30 ++-
 drivers/net/ethernet/intel/ice/ice_vf_lib.c        |  16 +-
 .../net/ethernet/marvell/octeontx2/af/rvu_npc.c    |   1 -
 .../ethernet/mellanox/mlx5/core/en_accel/macsec.c  |   1 +
 drivers/net/ethernet/mellanox/mlxsw/core.c         |   2 +-
 .../mellanox/mlxsw/core_acl_flex_actions.c         |  16 +-
 .../ethernet/mellanox/mlxsw/core_acl_flex_keys.c   |   9 +-
 drivers/net/ethernet/mellanox/mlxsw/core_env.c     |  20 +-
 drivers/net/ethernet/mellanox/mlxsw/pci.c          |  10 +-
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c |  11 +-
 .../ethernet/mellanox/mlxsw/spectrum_acl_tcam.c    | 132 ++++++----
 .../ethernet/mellanox/mlxsw/spectrum_acl_tcam.h    |   5 +-
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  |  15 +-
 .../ethernet/mellanox/mlxsw/spectrum_switchdev.c   |   8 +-
 drivers/net/ethernet/ti/am65-cpts.c                |   5 +
 drivers/net/ethernet/ti/icssg/icssg_prueth.c       |   8 +-
 drivers/net/ethernet/wangxun/libwx/wx_lib.c        |   2 +-
 drivers/net/gtp.c                                  |   3 +-
 drivers/net/macsec.c                               |  46 +++-
 drivers/net/phy/dp83869.c                          |   3 +-
 drivers/net/phy/mediatek-ge-soc.c                  |  43 ++--
 drivers/net/usb/ax88179_178a.c                     |  11 +-
 drivers/net/vxlan/vxlan_core.c                     |   4 +
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c |   2 +
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c      |   3 +-
 drivers/net/wireless/virtual/mac80211_hwsim.c      |   2 +-
 drivers/nfc/trf7970a.c                             |  42 ++--
 drivers/phy/freescale/phy-fsl-imx8m-pcie.c         |   6 +-
 drivers/phy/marvell/phy-mvebu-a3700-comphy.c       |   9 +-
 drivers/phy/qualcomm/phy-qcom-m31.c                |   2 +-
 drivers/phy/qualcomm/phy-qcom-qmp-combo.c          |  12 +-
 drivers/phy/qualcomm/phy-qcom-qmp.h                |   2 +
 drivers/phy/rockchip/phy-rockchip-naneng-combphy.c |  36 ++-
 drivers/phy/rockchip/phy-rockchip-snps-pcie3.c     |  31 +--
 drivers/phy/ti/phy-tusb1210.c                      |  23 +-
 drivers/soc/mediatek/mtk-svs.c                     |   7 +-
 drivers/soundwire/amd_manager.c                    |  15 ++
 drivers/soundwire/amd_manager.h                    |   3 +-
 drivers/usb/host/xhci-ring.c                       | 116 +++++----
 drivers/video/fbdev/core/fb_defio.c                |   2 +-
 fs/btrfs/backref.c                                 |  12 +-
 fs/btrfs/extent_map.c                              |   2 +-
 fs/btrfs/inode.c                                   |  13 +-
 fs/btrfs/scrub.c                                   |  18 +-
 fs/btrfs/tests/extent-map-tests.c                  |   5 +
 fs/netfs/buffered_write.c                          |  23 +-
 fs/proc/page.c                                     |   7 +-
 fs/smb/client/cifsfs.c                             |   1 +
 fs/smb/client/cifsglob.h                           |   2 +
 fs/smb/client/cifspdu.h                            |   4 +-
 fs/smb/client/fs_context.c                         |  12 +
 fs/smb/client/fs_context.h                         |   2 +
 fs/smb/client/fscache.c                            |  13 +
 fs/smb/client/misc.c                               |   3 +
 fs/smb/client/smb2pdu.h                            |   2 +-
 fs/smb/client/transport.c                          |   7 +-
 include/drm/drm_gem.h                              |  13 +
 include/linux/etherdevice.h                        |  25 ++
 include/linux/firmware/qcom/qcom_qseecom.h         |  55 ++++-
 include/linux/firmware/qcom/qcom_scm.h             |  10 +-
 include/linux/mm.h                                 |   8 +-
 include/linux/page-flags.h                         | 146 ++++++-----
 include/net/af_unix.h                              |   3 +
 include/net/bluetooth/hci.h                        |   8 +
 include/net/bluetooth/hci_core.h                   |  38 ++-
 include/net/bluetooth/hci_sync.h                   |  20 +-
 include/net/bluetooth/l2cap.h                      |   2 +-
 include/net/macsec.h                               |   2 +
 include/net/sock.h                                 |  41 +--
 include/net/tls.h                                  |   3 +-
 include/trace/events/mmflags.h                     |   1 +
 init/Kconfig                                       |   2 +-
 kernel/bounds.c                                    |   2 +-
 kernel/cpu.c                                       |   4 +-
 kernel/crash_core.c                                |   5 +-
 kernel/sched/fair.c                                |  34 +--
 lib/stackdepot.c                                   |   4 +-
 mm/hugetlb.c                                       |  29 +--
 mm/zswap.c                                         |  25 +-
 net/ax25/af_ax25.c                                 |   2 +-
 net/bluetooth/6lowpan.c                            |   2 +-
 net/bluetooth/hci_conn.c                           | 168 +++----------
 net/bluetooth/hci_event.c                          |  52 ++--
 net/bluetooth/hci_sync.c                           | 274 ++++++++++++++++++++-
 net/bluetooth/iso.c                                |  50 +++-
 net/bluetooth/l2cap_core.c                         |  12 +-
 net/bluetooth/l2cap_sock.c                         |  10 +-
 net/bluetooth/mgmt.c                               |  27 +-
 net/bluetooth/sco.c                                |  10 +-
 net/bridge/br_netlink.c                            |   2 +-
 net/core/sock.c                                    |   1 +
 net/core/sysctl_net_core.c                         |   9 +
 net/ethernet/eth.c                                 |  12 +-
 net/ipv4/icmp.c                                    |  12 +-
 net/ipv4/route.c                                   |   3 +
 net/ipv4/tcp_ao.c                                  |   3 +-
 net/ipv4/udp.c                                     |   5 +-
 net/ipv6/udp.c                                     |   5 +-
 net/mac80211/mesh.c                                |   8 +-
 net/mac80211/mesh.h                                |  36 ++-
 net/mac80211/mesh_pathtbl.c                        |  37 +--
 net/mac80211/mlme.c                                |   9 +-
 net/mac80211/rx.c                                  |  13 +-
 net/netfilter/ipvs/ip_vs_proto_sctp.c              |   6 +-
 net/netfilter/nft_chain_filter.c                   |   4 +-
 net/openvswitch/conntrack.c                        |   4 +-
 net/tls/tls.h                                      |   2 +-
 net/tls/tls_strp.c                                 |   6 +-
 net/unix/garbage.c                                 |   2 +-
 rust/Makefile                                      |   1 -
 rust/kernel/init.rs                                |  11 +-
 rust/kernel/lib.rs                                 |   2 +-
 rust/kernel/net/phy.rs                             |   4 +
 rust/macros/lib.rs                                 |  12 -
 scripts/Makefile.build                             |   2 +-
 tools/net/ynl/lib/ynl.py                           |   1 +
 tools/testing/selftests/riscv/hwprobe/cbo.c        |   2 +-
 tools/testing/selftests/riscv/hwprobe/hwprobe.h    |  10 +
 tools/testing/selftests/seccomp/seccomp_bpf.c      |  41 ++-
 229 files changed, 2425 insertions(+), 1546 deletions(-)
author maintainer
#1 lore raw

PATCH 1/228 cifs: Fix reacquisition of volume cookie on still-live connection

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Howells <[email protected]>

[ Upstream commit dad80c6bff770d25f67ec25fe011730e4a463008 ]

During mount, cifs_mount_get_tcon() gets a tcon resource connection record and then attaches an fscache volume cookie to it. However, it does this irrespective of whether or not the tcon returned from cifs_get_tcon() is a new record or one that's already in use. This leads to a warning about a volume cookie collision and a leaked volume cookie because tcon->fscache gets reset.

Fix this be adding a mutex and a "we've already tried this" flag and only doing it once for the lifetime of the tcon.

[!] Note: Looking at cifs_mount_get_tcon(), a more general solution may actually be required. Reacquiring the volume cookie isn't the only thing that function does: it also partially reinitialises the tcon record without any locking - which may cause live filesystem ops already using the tcon through a previous mount to malfunction.

This can be reproduced simply by something like:

mount //example.com/test /xfstest.test -o user=shares,pass=xxx,fsc mount //example.com/test /mnt -o user=shares,pass=xxx,fsc

Fixes 70431bfd825d ("cifs: Support fscache indexing rewrite")
cc Shyam Prasad N <[email protected]>
---
 fs/smb/client/cifsglob.h |  2 ++
 fs/smb/client/fscache.c  | 13 +++++++++++++
 fs/smb/client/misc.c     |  3 +++
 3 files changed, 18 insertions(+)
fs/smb/client/cifsglob.h +2 -0
index 844afda090d05..0c3311de5dc0a 100644
@@ -1261,7 +1261,9 @@ struct cifs_tcon {
 	__u32 max_cached_dirs;
 #ifdef CONFIG_CIFS_FSCACHE
 	u64 resource_id;		/* server resource id */
+	bool fscache_acquired;		/* T if we've tried acquiring a cookie */
 	struct fscache_volume *fscache;	/* cookie for share */
+	struct mutex fscache_lock;	/* Prevent regetting a cookie */
 #endif
 	struct list_head pending_opens;	/* list of incomplete opens */
 	struct cached_fids *cfids;
fs/smb/client/fscache.c +13 -0
index 340efce8f0529..113bde8f1e613 100644
@@ -43,12 +43,23 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon)
 	char *key;
 	int ret = -ENOMEM;
 
+	if (tcon->fscache_acquired)
+		return 0;
+
+	mutex_lock(&tcon->fscache_lock);
+	if (tcon->fscache_acquired) {
+		mutex_unlock(&tcon->fscache_lock);
+		return 0;
+	}
+	tcon->fscache_acquired = true;
+
 	tcon->fscache = NULL;
 	switch (sa->sa_family) {
 	case AF_INET:
 	case AF_INET6:
 		break;
 	default:
+		mutex_unlock(&tcon->fscache_lock);
 		cifs_dbg(VFS, "Unknown network family '%d'\n", sa->sa_family);
 		return -EINVAL;
 	}
@@ -57,6 +68,7 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon)
 
 	sharename = extract_sharename(tcon->tree_name);
 	if (IS_ERR(sharename)) {
+		mutex_unlock(&tcon->fscache_lock);
 		cifs_dbg(FYI, "%s: couldn't extract sharename\n", __func__);
 		return PTR_ERR(sharename);
 	}
@@ -90,6 +102,7 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon)
 	kfree(key);
 out:
 	kfree(sharename);
+	mutex_unlock(&tcon->fscache_lock);
 	return ret;
 }
 
fs/smb/client/misc.c +3 -0
index 74627d647818a..0d13db80e67c9 100644
@@ -141,6 +141,9 @@ tcon_info_alloc(bool dir_leases_enabled)
 	atomic_set(&ret_buf->num_local_opens, 0);
 	atomic_set(&ret_buf->num_remote_opens, 0);
 	ret_buf->stats_from_time = ktime_get_real_seconds();
+#ifdef CONFIG_CIFS_FSCACHE
+	mutex_init(&ret_buf->fscache_lock);
+#endif
 
 	return ret_buf;
 }
Signature · 1 line
2.43.0
author maintainer
#2 lore raw

PATCH 2/228 smb: client: fix rename(2) regression against samba

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paulo Alcantara <[email protected]>

[ Upstream commit 18d86965e31f9be4d477da0744a7cdc9815858de ]

After commit 2c7d399e551c ("smb: client: reuse file lease key in compound operations") the client started reusing lease keys for rename, unlink and set path size operations to prevent it from breaking its own leases and thus causing unnecessary lease breaks to same connection.

The implementation relies on positive dentries and cifsInodeInfo::lease_granted to decide whether reusing lease keys for the compound requests. cifsInodeInfo::lease_granted was introduced by commit 0ab95c2510b6 ("Defer close only when lease is enabled.") to indicate whether lease caching is granted for a specific file, but that can only happen until file is open, so cifsInodeInfo::lease_granted was left uninitialised in ->alloc_inode and then client started sending random lease keys for files that hadn't any leases.

This fixes the following test case against samba:

mount.cifs //srv/share /mnt/1 -o ...,nosharesock
mount.cifs //srv/share /mnt/2 -o ...,nosharesock
touch /mnt/1/foo; tail -f /mnt/1/foo & pid=$!
mv /mnt/2/foo /mnt/2/bar # fails with -EIO
kill $pid

Fixes 0ab95c2510b6 ("Defer close only when lease is enabled.")
---
 fs/smb/client/cifsfs.c | 1 +
 1 file changed, 1 insertion(+)
fs/smb/client/cifsfs.c +1 -0
index 501bfd3afdc64..81fe60815885c 100644
@@ -393,6 +393,7 @@ cifs_alloc_inode(struct super_block *sb)
 	 * server, can not assume caching of file data or metadata.
 	 */
 	cifs_set_oplock_level(cifs_inode, 0);
+	cifs_inode->lease_granted = false;
 	cifs_inode->flags = 0;
 	spin_lock_init(&cifs_inode->writers_lock);
 	cifs_inode->writers = 0;
Signature · 1 line
2.43.0
author maintainer
#3 lore raw

PATCH 3/228 cifs: reinstate original behavior again for forceuid/forcegid

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Takayuki Nagata <[email protected]>

[ Upstream commit 77d8aa79ecfb209308e0644c02f655122b31def7 ]

forceuid/forcegid should be enabled by default when uid=/gid= options are specified, but commit 24e0a1eff9e2 ("cifs: switch to new mount api") changed the behavior. Due to the change, a mounted share does not show intentional uid/gid for files and directories even though uid=/gid= options are specified since forceuid/forcegid are not enabled.

This patch reinstates original behavior that overrides uid/gid with specified uid/gid by the options.

---
 fs/smb/client/fs_context.c | 12 ++++++++++++
 fs/smb/client/fs_context.h |  2 ++
 2 files changed, 14 insertions(+)
fs/smb/client/fs_context.c +12 -0
index d9fe17b2ba375..775b0ca605892 100644
@@ -715,6 +715,16 @@ static int smb3_fs_context_validate(struct fs_context *fc)
 	/* set the port that we got earlier */
 	cifs_set_port((struct sockaddr *)&ctx->dstaddr, ctx->port);
 
+	if (ctx->uid_specified && !ctx->forceuid_specified) {
+		ctx->override_uid = 1;
+		pr_notice("enabling forceuid mount option implicitly because uid= option is specified\n");
+	}
+
+	if (ctx->gid_specified && !ctx->forcegid_specified) {
+		ctx->override_gid = 1;
+		pr_notice("enabling forcegid mount option implicitly because gid= option is specified\n");
+	}
+
 	if (ctx->override_uid && !ctx->uid_specified) {
 		ctx->override_uid = 0;
 		pr_notice("ignoring forceuid mount option specified with no uid= option\n");
@@ -984,12 +994,14 @@ static int smb3_fs_context_parse_param(struct fs_context *fc,
 			ctx->override_uid = 0;
 		else
 			ctx->override_uid = 1;
+		ctx->forceuid_specified = true;
 		break;
 	case Opt_forcegid:
 		if (result.negated)
 			ctx->override_gid = 0;
 		else
 			ctx->override_gid = 1;
+		ctx->forcegid_specified = true;
 		break;
 	case Opt_perm:
 		if (result.negated)
fs/smb/client/fs_context.h +2 -0
index b3d51b345eaef..61776572b8d2d 100644
@@ -156,6 +156,8 @@ enum cifs_param {
 };
 
 struct smb3_fs_context {
+	bool forceuid_specified;
+	bool forcegid_specified;
 	bool uid_specified;
 	bool cruid_specified;
 	bool gid_specified;
Signature · 1 line
2.43.0
author maintainer
#4 lore raw

PATCH 4/228 HID: intel-ish-hid: ipc: Fix dev_err usage with uninitialized dev->devc

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhang Lixu <[email protected]>

[ Upstream commit 92826905ae340b7f2b25759a06c8c60bfc476b9f ]

The variable dev->devc in ish_dev_init was utilized by dev_err before it was properly assigned. To rectify this, the assignment of dev->devc has been moved to immediately follow memory allocation.

Without this change "(NULL device *)" is printed for device information.

Fixes 8ae2f2b0a284 ("HID: intel-ish-hid: ipc: Fix potential use-after-free in work function")
Fixes ae02e5d40d5f ("HID: intel-ish-hid: ipc layer")
---
 drivers/hid/intel-ish-hid/ipc/ipc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hid/intel-ish-hid/ipc/ipc.c +1 -1
index a49c6affd7c4c..dd5fc60874ba1 100644
@@ -948,6 +948,7 @@ struct ishtp_device *ish_dev_init(struct pci_dev *pdev)
 	if (!dev)
 		return NULL;
 
+	dev->devc = &pdev->dev;
 	ishtp_device_init(dev);
 
 	init_waitqueue_head(&dev->wait_hw_ready);
@@ -983,7 +984,6 @@ struct ishtp_device *ish_dev_init(struct pci_dev *pdev)
 	}
 
 	dev->ops = &ish_hw_ops;
-	dev->devc = &pdev->dev;
 	dev->mtu = IPC_PAYLOAD_SIZE - sizeof(struct ishtp_msg_hdr);
 	return dev;
 }
Signature · 1 line
2.43.0
author maintainer
#5 lore raw

PATCH 5/228 HID: logitech-dj: allow mice to use all types of reports

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yaraslau Furman <[email protected]>

[ Upstream commit 21f28a7eb78dea6c59be6b0a5e0b47bf3d25fcbb ]

You can bind whatever action you want to the mouse's reprogrammable buttons using Windows application. Allow Linux to receive multimedia keycodes.

Fixes 3ed224e273ac ("HID: logitech-dj: Fix 064d:c52f receiver support")
---
 drivers/hid/hid-logitech-dj.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/hid/hid-logitech-dj.c +1 -3
index e6a8b6d8eab70..3c3c497b6b911 100644
@@ -965,9 +965,7 @@ static void logi_hidpp_dev_conn_notif_equad(struct hid_device *hdev,
 		}
 		break;
 	case REPORT_TYPE_MOUSE:
-		workitem->reports_supported |= STD_MOUSE | HIDPP;
-		if (djrcv_dev->type == recvr_type_mouse_only)
-			workitem->reports_supported |= MULTIMEDIA;
+		workitem->reports_supported |= STD_MOUSE | HIDPP | MULTIMEDIA;
 		break;
 	}
 }
Signature · 1 line
2.43.0
author maintainer
#6 lore raw

PATCH 6/228 arm64: dts: rockchip: set PHY address of MT7531 switch to 0x1f

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Arınç ÜNAL <[email protected]>

[ Upstream commit a2ac2a1b02590a22a236c43c455f421cdede45f5 ]

The MT7531 switch listens on PHY address 0x1f on an MDIO bus. I've got two findings that support this. There's no bootstrapping option to change the PHY address of the switch. The Linux driver hardcodes 0x1f as the PHY address of the switch. So the reg property on the device tree is currently ignored by the Linux driver.

Therefore, describe the correct PHY address on Banana Pi BPI-R2 Pro that has this switch.

---
 arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts +2 -2
index f9127ddfbb7df..87c45d8be420f 100644
@@ -525,9 +525,9 @@
 	#address-cells = <1>;
 	#size-cells = <0>;
 
-	switch@0 {
+	switch@1f {
 		compatible = "mediatek,mt7531";
-		reg = <0>;
+		reg = <0x1f>;
 
 		ports {
 			#address-cells = <1>;
Signature · 1 line
2.43.0
author maintainer
#7 lore raw

PATCH 7/228 arm64: dts: rockchip: enable internal pull-up on Q7_USB_ID for RK3399 Puma

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Quentin Schulz <[email protected]>

[ Upstream commit e6b1168f37e3f86d9966276c5a3fff9eb0df3e5f ]

The Q7_USB_ID has a diode used as a level-shifter, and is used as an input pin. The SoC default for this pin is a pull-up, which is correct but the pinconf in the introducing commit missed that, so let's fix this oversight.

Fixes ed2c66a95c0c ("arm64: dts: rockchip: fix rk3399-puma-haikou USB OTG mode")
---
 arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi +1 -1
index c08e69391c015..d4324e57729b5 100644
@@ -458,7 +458,7 @@
 	usb3 {
 		usb3_id: usb3-id {
 			rockchip,pins =
-			  <1 RK_PC2 RK_FUNC_GPIO &pcfg_pull_none>;
+			  <1 RK_PC2 RK_FUNC_GPIO &pcfg_pull_up>;
 		};
 	};
 };
Signature · 1 line
2.43.0
author maintainer
#8 lore raw

PATCH 8/228 arm64: dts: rockchip: fix alphabetical ordering RK3399 puma

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Iskander Amara <[email protected]>

[ Upstream commit f0abb4b2c7acf3c3e4130dc3f54cd90cf2ae62bc ]

Nodes overridden by their reference should be ordered alphabetically to make it easier to read the DTS. pinctrl node is defined in the wrong location so let's reorder it.

Stable-dep-of: 945a7c857091 ("arm64: dts: rockchip: enable internal pull-up on PCIE_WAKE# for RK3399 Puma")

---
 arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)
arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi +9 -9
index d4324e57729b5..a8102d1e31987 100644
@@ -416,15 +416,6 @@
 	gpio1830-supply = <&vcc_1v8>;
 };
 
-&pmu_io_domains {
-	status = "okay";
-	pmu1830-supply = <&vcc_1v8>;
-};
-
-&pwm2 {
-	status = "okay";
-};
-
 &pinctrl {
 	i2c8 {
 		i2c8_xfer_a: i2c8-xfer {
@@ -463,6 +454,15 @@
 	};
 };
 
+&pmu_io_domains {
+	status = "okay";
+	pmu1830-supply = <&vcc_1v8>;
+};
+
+&pwm2 {
+	status = "okay";
+};
+
 &sdhci {
 	/*
 	 * Signal integrity isn't great at 200MHz but 100MHz has proven stable
Signature · 1 line
2.43.0
author maintainer
#9 lore raw

PATCH 9/228 arm64: dts: rockchip: enable internal pull-up on PCIE_WAKE# for RK3399 Puma

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Quentin Schulz <[email protected]>

[ Upstream commit 945a7c8570916650a415757d15d83e0fa856a686 ]

The PCIE_WAKE# has a diode used as a level-shifter, and is used as an input pin. While the SoC default is to enable the pull-up, the core rk3399 pinconf for this pin opted for pull-none. So as to not disturb the behaviour of other boards which may rely on pull-none instead of pull-up, set the needed pull-up only for RK3399 Puma.

Fixes 60fd9f72ce8a ("arm64: dts: rockchip: add Haikou baseboard with RK3399-Q7 SoM")
---
 arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi | 5 +++++
 1 file changed, 5 insertions(+)
arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi +5 -0
index a8102d1e31987..c0eb9f39c33e5 100644
@@ -416,6 +416,11 @@
 	gpio1830-supply = <&vcc_1v8>;
 };
 
+&pcie_clkreqn_cpm {
+	rockchip,pins =
+		<2 RK_PD2 RK_FUNC_GPIO &pcfg_pull_up>;
+};
+
 &pinctrl {
 	i2c8 {
 		i2c8_xfer_a: i2c8-xfer {
Signature · 1 line
2.43.0
author maintainer
#10 lore raw

PATCH 10/228 arm64: dts: rockchip: Fix the i2c address of es8316 on Cool Pi CM5

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Yan <[email protected]>

[ Upstream commit 64da060dd4eb625646970d7c96a16de617412ec5 ]

According to the hardware design, the i2c address of audio codec es8316 on Cool Pi CM5 is 0x10.

This fix the read/write error like bellow:
es8316 7-0011: ASoC: error at soc_component_write_no_lock on es8316.7-0011 for register: [0x0000000c] -6 es8316 7-0011: ASoC: error at soc_component_write_no_lock on es8316.7-0011 for register: [0x00000003] -6 es8316 7-0011: ASoC: error at soc_component_read_no_lock on es8316.7-0011 for register: [0x00000016] -6 es8316 7-0011: ASoC: error at soc_component_read_no_lock on es8316.7-0011 for register: [0x00000016] -6

Fixes 791c154c3982 ("arm64: dts: rockchip: Add support for rk3588 based board Cool Pi CM5 EVB")
[also adapted the node name to audio-codec@10]
---
 arch/arm64/boot/dts/rockchip/rk3588-coolpi-cm5.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/rockchip/rk3588-coolpi-cm5.dtsi +2 -2
index cce1c8e835877..94ecb9b4f98f8 100644
@@ -216,9 +216,9 @@
 	pinctrl-0 = <&i2c7m0_xfer>;
 	status = "okay";
 
-	es8316: audio-codec@11 {
+	es8316: audio-codec@10 {
 		compatible = "everest,es8316";
-		reg = <0x11>;
+		reg = <0x10>;
 		assigned-clocks = <&cru I2S0_8CH_MCLKOUT>;
 		assigned-clock-rates = <12288000>;
 		clocks = <&cru I2S0_8CH_MCLKOUT>;
Signature · 1 line
2.43.0
author maintainer
#11 lore raw

PATCH 11/228 arm64: dts: rockchip: Remove unsupported node from the Pinebook Pro dts

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dragan Simic <[email protected]>

[ Upstream commit 43853e843aa6c3d47ff2b0cce898318839483d05 ]

Remove a redundant node from the Pine64 Pinebook Pro dts, which is intended to provide a value for the delay in PCI Express enumeration, but that isn't supported without additional out-of-tree kernel patches.

There were already efforts to upstream those kernel patches, because they reportedly make some PCI Express cards (such as LSI SAS HBAs) usable in Pine64 RockPro64 (which is also based on the RK3399); otherwise, those PCI Express cards fail to enumerate. However, providing the required background and explanations proved to be a tough nut to crack, which is the reason why those patches remain outside of the kernel mainline for now.

If those out-of-tree patches eventually become upstreamed, the resulting device-tree changes will almost surely belong to the RK3399 SoC dtsi. Also, the above-mentioned unusable-without-out-of-tree-patches PCI Express devices are in all fairness not usable in a Pinebook Pro without some extensive hardware modifications, which is another reason to delete this redundant node. When it comes to the Pinebook Pro, only M.2 NVMe SSDs can be installed out of the box (using an additional passive adapter PCB sold separately by Pine64), which reportedly works fine with no additional patches.

---
 arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts | 1 -
 1 file changed, 1 deletion(-)
arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts +0 -1
index 054c6a4d1a45f..294eb2de263de 100644
@@ -779,7 +779,6 @@
 };
 
 &pcie0 {
-	bus-scan-delay-ms = <1000>;
 	ep-gpios = <&gpio2 RK_PD4 GPIO_ACTIVE_HIGH>;
 	num-lanes = <4>;
 	pinctrl-names = "default";
Signature · 1 line
2.43.0
author maintainer
#12 lore raw

PATCH 12/228 arm64: dts: mediatek: mt8183: Add power-domains properity to mfgcfg

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ikjoon Jang <[email protected]>

[ Upstream commit 1781f2c461804c0123f59afc7350e520a88edffb ]

mfgcfg clock is under MFG_ASYNC power domain.

Fixes e526c9bc11f8 ("arm64: dts: Add Mediatek SoC MT8183 and evaluation board dts and Makefile")
Fixes 37fb78b9aeb7 ("arm64: dts: mediatek: Add mt8183 power domains controller")
---
 arch/arm64/boot/dts/mediatek/mt8183.dtsi | 1 +
 1 file changed, 1 insertion(+)
arch/arm64/boot/dts/mediatek/mt8183.dtsi +1 -0
index 920ee415ef5fb..e1767a4e38f3a 100644
@@ -1628,6 +1628,7 @@
 			compatible = "mediatek,mt8183-mfgcfg", "syscon";
 			reg = <0 0x13000000 0 0x1000>;
 			#clock-cells = <1>;
+			power-domains = <&spm MT8183_POWER_DOMAIN_MFG_ASYNC>;
 		};
 
 		gpu: gpu@13040000 {
Signature · 1 line
2.43.0
author maintainer
#13 lore raw

PATCH 13/228 arm64: dts: mediatek: mt8192: Add missing gce-client-reg to mutex

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 00bcc8810d9dd69d3899a4189e2f3964f263a600 ]

Add the missing mediatek,gce-client-reg property to the mutex node to allow it to use the GCE. This prevents the "can't parse gce-client-reg property" error from being printed and should result in better performance.

---
 arch/arm64/boot/dts/mediatek/mt8192.dtsi | 1 +
 1 file changed, 1 insertion(+)
arch/arm64/boot/dts/mediatek/mt8192.dtsi +1 -0
index 0e432636b8c23..eea8d141f93cf 100644
@@ -1456,6 +1456,7 @@
 			reg = <0 0x14001000 0 0x1000>;
 			interrupts = <GIC_SPI 252 IRQ_TYPE_LEVEL_HIGH 0>;
 			clocks = <&mmsys CLK_MM_DISP_MUTEX0>;
+			mediatek,gce-client-reg = <&gce SUBSYS_1400XXXX 0x1000 0x1000>;
 			mediatek,gce-events = <CMDQ_EVENT_DISP_STREAM_DONE_ENG_EVENT_0>,
 					      <CMDQ_EVENT_DISP_STREAM_DONE_ENG_EVENT_1>;
 			power-domains = <&spm MT8192_POWER_DOMAIN_DISP>;
Signature · 1 line
2.43.0
author maintainer
#14 lore raw

PATCH 14/228 arm64: dts: mediatek: mt8195: Add missing gce-client-reg to vpp/vdosys

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 96b0c1528ef41fe754f5d1378b1db6c098a2e33f ]

Add the missing mediatek,gce-client-reg property to the vppsys and vdosys nodes to allow them to use the GCE. This prevents the "can't parse gce-client-reg property" error from being printed and should result in better performance.

---
 arch/arm64/boot/dts/mediatek/mt8195.dtsi | 3 +++
 1 file changed, 3 insertions(+)
arch/arm64/boot/dts/mediatek/mt8195.dtsi +3 -0
index b9101662ce40d..775aeeb6ebd37 100644
@@ -1998,6 +1998,7 @@
 			compatible = "mediatek,mt8195-vppsys0", "syscon";
 			reg = <0 0x14000000 0 0x1000>;
 			#clock-cells = <1>;
+			mediatek,gce-client-reg = <&gce1 SUBSYS_1400XXXX 0 0x1000>;
 		};
 
 		dma-controller@14001000 {
@@ -2221,6 +2222,7 @@
 			compatible = "mediatek,mt8195-vppsys1", "syscon";
 			reg = <0 0x14f00000 0 0x1000>;
 			#clock-cells = <1>;
+			mediatek,gce-client-reg = <&gce1 SUBSYS_14f0XXXX 0 0x1000>;
 		};
 
 		mutex@14f01000 {
@@ -3050,6 +3052,7 @@
 			reg = <0 0x1c01a000 0 0x1000>;
 			mboxes = <&gce0 0 CMDQ_THR_PRIO_4>;
 			#clock-cells = <1>;
+			mediatek,gce-client-reg = <&gce0 SUBSYS_1c01XXXX 0xa000 0x1000>;
 		};
 
 
Signature · 1 line
2.43.0
author maintainer
#15 lore raw

PATCH 15/228 arm64: dts: mediatek: mt8195: Add missing gce-client-reg to mutex

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 3b129949184a1251e6a42db714f6d68b75fabedd ]

Add the missing mediatek,gce-client-reg property to the mutex node to allow it to use the GCE. This prevents the "can't parse gce-client-reg property" error from being printed and should result in better performance.

---
 arch/arm64/boot/dts/mediatek/mt8195.dtsi | 1 +
 1 file changed, 1 insertion(+)
arch/arm64/boot/dts/mediatek/mt8195.dtsi +1 -0
index 775aeeb6ebd37..6cea6612e2feb 100644
@@ -3234,6 +3234,7 @@
 			interrupts = <GIC_SPI 658 IRQ_TYPE_LEVEL_HIGH 0>;
 			power-domains = <&spm MT8195_POWER_DOMAIN_VDOSYS0>;
 			clocks = <&vdosys0 CLK_VDO0_DISP_MUTEX0>;
+			mediatek,gce-client-reg = <&gce0 SUBSYS_1c01XXXX 0x6000 0x1000>;
 			mediatek,gce-events = <CMDQ_EVENT_VDO0_DISP_STREAM_DONE_0>;
 		};
 
Signature · 1 line
2.43.0
author maintainer
#16 lore raw

PATCH 16/228 arm64: dts: mediatek: mt8195: Add missing gce-client-reg to mutex1

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 58f126296c3c52d02bf3fad1f68c331d718c4a9b ]

Add the missing mediatek,gce-client-reg property to the mutex1 node to allow it to use the GCE. This prevents the "can't parse gce-client-reg property" error from being printed and should result in better performance.

---
 arch/arm64/boot/dts/mediatek/mt8195.dtsi | 1 +
 1 file changed, 1 insertion(+)
arch/arm64/boot/dts/mediatek/mt8195.dtsi +1 -0
index 6cea6612e2feb..4dd7f755630ea 100644
@@ -3305,6 +3305,7 @@
 			power-domains = <&spm MT8195_POWER_DOMAIN_VDOSYS1>;
 			clocks = <&vdosys1 CLK_VDO1_DISP_MUTEX>;
 			clock-names = "vdo1_mutex";
+			mediatek,gce-client-reg = <&gce0 SUBSYS_1c10XXXX 0x1000 0x1000>;
 			mediatek,gce-events = <CMDQ_EVENT_VDO1_STREAM_DONE_ENG_0>;
 		};
 
Signature · 1 line
2.43.0
author maintainer
#17 lore raw

PATCH 17/228 arm64: dts: mediatek: cherry: Describe CPU supplies

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 17b33dd9e4a38fbaca87c68e532b52f9d0492ba7 ]

Describe in each CPU node the regulator supplying it.

Fixes 260c04d425eb ("arm64: dts: mediatek: cherry: Enable MT6315 regulators on SPMI bus")
---
 .../boot/dts/mediatek/mt8195-cherry.dtsi      | 32 +++++++++++++++++++
 1 file changed, 32 insertions(+)
arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi +32 -0
index 3c6079edda190..1f3a6755f1550 100644
@@ -264,6 +264,38 @@
 	status = "okay";
 };
 
+&cpu0 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu1 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu2 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu3 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu4 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
+&cpu5 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
+&cpu6 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
+&cpu7 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
 &dp_intf0 {
 	status = "okay";
 
Signature · 1 line
2.43.0
author maintainer
#18 lore raw

PATCH 18/228 arm64: dts: mediatek: mt8192-asurada: Update min voltage constraint for MT6315

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pin-yen Lin <[email protected]>

[ Upstream commit 374a7c6400e314458178255a63c37d6347845092 ]

Update the minimum voltage from 300000 uV to 400000 uV so it matches the MT6315 datasheet.

Also update the minimum voltage for Vgpu regulator from 606250 uV to 400000 uV because the requested voltage could be lower than the minimum voltage on the GPU OPP table when the MTK Smart Voltage Scaling (SVS) driver is enabled.

---
 arch/arm64/boot/dts/mediatek/mt8192-asurada.dtsi | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/mediatek/mt8192-asurada.dtsi +3 -3
index d4dd5e8b2e1d3..f62b2498eb1c7 100644
@@ -1420,7 +1420,7 @@
 			mt6315_6_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vbcpu";
-				regulator-min-microvolt = <300000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-allowed-modes = <0 1 2>;
@@ -1430,7 +1430,7 @@
 			mt6315_6_vbuck3: vbuck3 {
 				regulator-compatible = "vbuck3";
 				regulator-name = "Vlcpu";
-				regulator-min-microvolt = <300000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-allowed-modes = <0 1 2>;
@@ -1447,7 +1447,7 @@
 			mt6315_7_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vgpu";
-				regulator-min-microvolt = <606250>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <800000>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-allowed-modes = <0 1 2>;
Signature · 1 line
2.43.0
author maintainer
#19 lore raw

PATCH 19/228 arm64: dts: mediatek: mt8195-cherry: Update min voltage constraint for MT6315

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pin-yen Lin <[email protected]>

[ Upstream commit e9a6b8b5c61350535c7eb5ea9b2dde0d5745bd1b ]

Update the minimum voltage from 300000 uV to 400000 uV so it matches the MT6315 datasheet.

Also update the minimum voltage for Vgpu regulator from 625000 uV to 400000 uV because the requested voltage could be lower than the minimum voltage on the GPU OPP table when the MTK Smart Voltage Scaling (SVS) driver is enabled.

Fixes 260c04d425eb ("arm64: dts: mediatek: cherry: Enable MT6315 regulators on SPMI bus")
---
 arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi +2 -2
index 1f3a6755f1550..121dfd91da07b 100644
@@ -1245,7 +1245,7 @@
 			mt6315_6_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vbcpu";
-				regulator-min-microvolt = <300000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-ramp-delay = <6250>;
@@ -1263,7 +1263,7 @@
 			mt6315_7_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vgpu";
-				regulator-min-microvolt = <625000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-ramp-delay = <6250>;
Signature · 1 line
2.43.0
author maintainer
#20 lore raw

PATCH 20/228 arm64: dts: mediatek: mt8183-kukui: Use default min voltage for MT6358

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pin-yen Lin <[email protected]>

[ Upstream commit 296118a8dc297de47d9b3a364b9743f8446bd612 ]

The requested voltage could be lower than the minimum voltage on the GPU OPP table when the MTK Smart Voltage Scaling (SVS) driver is enabled, so removing the definition in mt8183-kukui to use the default minimum voltage (500000 uV) defined in mt6358.dtsi.

Fixes 31c6732da9d5 ("arm64: dts: mediatek: mt8183-kukui: Override vgpu/vsram_gpu constraints")
---
 arch/arm64/boot/dts/mediatek/mt8183-kukui.dtsi | 1 -
 1 file changed, 1 deletion(-)
arch/arm64/boot/dts/mediatek/mt8183-kukui.dtsi +0 -1
index 90c5ad917a9ba..41aef6c3edfc5 100644
@@ -433,7 +433,6 @@
 };
 
 &mt6358_vgpu_reg {
-	regulator-min-microvolt = <625000>;
 	regulator-max-microvolt = <900000>;
 
 	regulator-coupled-with = <&mt6358_vsram_gpu_reg>;
Signature · 1 line
2.43.0
author maintainer
#21 lore raw

PATCH 21/228 arm64: dts: mediatek: mt7622: fix clock controllers

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3ba5a61594347ab46e7c2cff6cd63ea0f1282efb ]

1. Drop unneeded "syscon"s (bindings were updated recently)
2. Use "clock-controller" in nodenames
3. Add missing "#clock-cells"

Fixes d7167881e03e ("arm64: dts: mt7622: add clock controller device nodes")
Fixes e9b65ecb7c30 ("arm64: dts: mediatek: mt7622: introduce nodes for Wireless Ethernet Dispatch")
---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 27 +++++++++++-------------
 1 file changed, 12 insertions(+), 15 deletions(-)
arch/arm64/boot/dts/mediatek/mt7622.dtsi +12 -15
index 3ee9266fa8e98..283fdf7d2d8b9 100644
@@ -283,16 +283,14 @@
 		};
 	};
 
-	apmixedsys: apmixedsys@10209000 {
-		compatible = "mediatek,mt7622-apmixedsys",
-			     "syscon";
+	apmixedsys: clock-controller@10209000 {
+		compatible = "mediatek,mt7622-apmixedsys";
 		reg = <0 0x10209000 0 0x1000>;
 		#clock-cells = <1>;
 	};
 
-	topckgen: topckgen@10210000 {
-		compatible = "mediatek,mt7622-topckgen",
-			     "syscon";
+	topckgen: clock-controller@10210000 {
+		compatible = "mediatek,mt7622-topckgen";
 		reg = <0 0x10210000 0 0x1000>;
 		#clock-cells = <1>;
 	};
@@ -734,9 +732,8 @@
 		power-domains = <&scpsys MT7622_POWER_DOMAIN_WB>;
 	};
 
-	ssusbsys: ssusbsys@1a000000 {
-		compatible = "mediatek,mt7622-ssusbsys",
-			     "syscon";
+	ssusbsys: clock-controller@1a000000 {
+		compatible = "mediatek,mt7622-ssusbsys";
 		reg = <0 0x1a000000 0 0x1000>;
 		#clock-cells = <1>;
 		#reset-cells = <1>;
@@ -793,9 +790,8 @@
 		};
 	};
 
-	pciesys: pciesys@1a100800 {
-		compatible = "mediatek,mt7622-pciesys",
-			     "syscon";
+	pciesys: clock-controller@1a100800 {
+		compatible = "mediatek,mt7622-pciesys";
 		reg = <0 0x1a100800 0 0x1000>;
 		#clock-cells = <1>;
 		#reset-cells = <1>;
@@ -921,12 +917,13 @@
 		};
 	};
 
-	hifsys: syscon@1af00000 {
-		compatible = "mediatek,mt7622-hifsys", "syscon";
+	hifsys: clock-controller@1af00000 {
+		compatible = "mediatek,mt7622-hifsys";
 		reg = <0 0x1af00000 0 0x70>;
+		#clock-cells = <1>;
 	};
 
-	ethsys: syscon@1b000000 {
+	ethsys: clock-controller@1b000000 {
 		compatible = "mediatek,mt7622-ethsys",
 			     "syscon";
 		reg = <0 0x1b000000 0 0x1000>;
Signature · 1 line
2.43.0
author maintainer
#22 lore raw

PATCH 22/228 arm64: dts: mediatek: mt7622: fix IR nodename

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 800dc93c3941e372c94278bf4059e6e82f60bd66 ]

Fix following validation error:
arch/arm64/boot/dts/mediatek/mt7622-rfb1.dtb: cir@10009000: $nodename:0: 'cir@10009000' does not match '^ir(-receiver)?(@[a-f0-9]+)?$'
from schema $id: http://devicetree.org/schemas/media/mediatek,mt7622-cir.yaml#

Fixes ae457b7679c4 ("arm64: dts: mt7622: add SoC and peripheral related device nodes")
---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/mediatek/mt7622.dtsi +1 -1
index 283fdf7d2d8b9..4c8a71c8184b7 100644
@@ -252,7 +252,7 @@
 		clock-names = "hif_sel";
 	};
 
-	cir: cir@10009000 {
+	cir: ir-receiver@10009000 {
 		compatible = "mediatek,mt7622-cir";
 		reg = <0 0x10009000 0 0x1000>;
 		interrupts = <GIC_SPI 175 IRQ_TYPE_LEVEL_LOW>;
Signature · 1 line
2.43.0
author maintainer
#23 lore raw

PATCH 23/228 arm64: dts: mediatek: mt7622: fix ethernet controller "compatible"

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 208add29ce5b7291f6c466e4dfd9cbf61c72888e ]

Fix following validation error:
arch/arm64/boot/dts/mediatek/mt7622-rfb1.dtb: ethernet@1b100000: compatible: ['mediatek,mt7622-eth', 'mediatek,mt2701-eth', 'syscon'] is too long
from schema $id: http://devicetree.org/schemas/net/mediatek,net.yaml#
(and other complains about wrong clocks).

---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
arch/arm64/boot/dts/mediatek/mt7622.dtsi +1 -3
index 4c8a71c8184b7..8e46480b5364b 100644
@@ -963,9 +963,7 @@
 	};
 
 	eth: ethernet@1b100000 {
-		compatible = "mediatek,mt7622-eth",
-			     "mediatek,mt2701-eth",
-			     "syscon";
+		compatible = "mediatek,mt7622-eth";
 		reg = <0 0x1b100000 0 0x20000>;
 		interrupts = <GIC_SPI 223 IRQ_TYPE_LEVEL_LOW>,
 			     <GIC_SPI 224 IRQ_TYPE_LEVEL_LOW>,
Signature · 1 line
2.43.0
author maintainer
#24 lore raw

PATCH 24/228 arm64: dts: mediatek: mt7622: drop "reset-names" from thermal block

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit ecb5b0034f5bcc35003b4b965cf50c6e98316e79 ]

Binding doesn't specify "reset-names" property and Linux driver also doesn't use it.

Fix following validation error:
arch/arm64/boot/dts/mediatek/mt7622-rfb1.dtb: thermal@1100b000: Unevaluated properties are not allowed ('reset-names' was unexpected)
from schema $id: http://devicetree.org/schemas/thermal/mediatek,thermal.yaml#

Fixes ae457b7679c4 ("arm64: dts: mt7622: add SoC and peripheral related device nodes")
---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 1 -
 1 file changed, 1 deletion(-)
arch/arm64/boot/dts/mediatek/mt7622.dtsi +0 -1
index 8e46480b5364b..917fa39a74f8d 100644
@@ -513,7 +513,6 @@
 			 <&pericfg CLK_PERI_AUXADC_PD>;
 		clock-names = "therm", "auxadc";
 		resets = <&pericfg MT7622_PERI_THERM_SW_RST>;
-		reset-names = "therm";
 		mediatek,auxadc = <&auxadc>;
 		mediatek,apmixedsys = <&apmixedsys>;
 		nvmem-cells = <&thermal_calibration>;
Signature · 1 line
2.43.0
author maintainer
#25 lore raw

PATCH 25/228 arm64: dts: mediatek: mt7986: reorder properties

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 7eb133c99fbebc6adb1cbd22c926d42d2bbca648 ]

Use order described as preferred in DTS Coding Style. Mostly just move "compatible", "reg" and "ranges" properties. In two nodes also move vendor-prefixed props down.

Stable-dep-of: 3b449bfd2ff6 ("arm64: dts: mediatek: mt7986: drop invalid properties from ethsys")

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 67 ++++++++++++-----------
 1 file changed, 34 insertions(+), 33 deletions(-)
arch/arm64/boot/dts/mediatek/mt7986a.dtsi +34 -33
index d974739eae1c9..eba5e27a1bbea 100644
@@ -27,34 +27,34 @@
 		#address-cells = <1>;
 		#size-cells = <0>;
 		cpu0: cpu@0 {
-			device_type = "cpu";
 			compatible = "arm,cortex-a53";
-			enable-method = "psci";
 			reg = <0x0>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 
 		cpu1: cpu@1 {
-			device_type = "cpu";
 			compatible = "arm,cortex-a53";
-			enable-method = "psci";
 			reg = <0x1>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 
 		cpu2: cpu@2 {
-			device_type = "cpu";
 			compatible = "arm,cortex-a53";
-			enable-method = "psci";
 			reg = <0x2>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 
 		cpu3: cpu@3 {
-			device_type = "cpu";
-			enable-method = "psci";
 			compatible = "arm,cortex-a53";
 			reg = <0x3>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 	};
@@ -131,22 +131,22 @@
 	};
 
 	soc {
-		#address-cells = <2>;
-		#size-cells = <2>;
 		compatible = "simple-bus";
 		ranges;
+		#address-cells = <2>;
+		#size-cells = <2>;
 
 		gic: interrupt-controller@c000000 {
 			compatible = "arm,gic-v3";
-			#interrupt-cells = <3>;
-			interrupt-parent = <&gic>;
-			interrupt-controller;
 			reg = <0 0x0c000000 0 0x10000>,  /* GICD */
 			      <0 0x0c080000 0 0x80000>,  /* GICR */
 			      <0 0x0c400000 0 0x2000>,   /* GICC */
 			      <0 0x0c410000 0 0x1000>,   /* GICH */
 			      <0 0x0c420000 0 0x2000>;   /* GICV */
+			interrupt-parent = <&gic>;
 			interrupts = <GIC_PPI 9 IRQ_TYPE_LEVEL_HIGH>;
+			interrupt-controller;
+			#interrupt-cells = <3>;
 		};
 
 		infracfg: infracfg@10001000 {
@@ -310,9 +310,9 @@
 
 		spi0: spi@1100a000 {
 			compatible = "mediatek,mt7986-spi-ipm", "mediatek,spi-ipm";
+			reg = <0 0x1100a000 0 0x100>;
 			#address-cells = <1>;
 			#size-cells = <0>;
-			reg = <0 0x1100a000 0 0x100>;
 			interrupts = <GIC_SPI 140 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&topckgen CLK_TOP_MPLL_D2>,
 				 <&topckgen CLK_TOP_SPI_SEL>,
@@ -324,9 +324,9 @@
 
 		spi1: spi@1100b000 {
 			compatible = "mediatek,mt7986-spi-ipm", "mediatek,spi-ipm";
+			reg = <0 0x1100b000 0 0x100>;
 			#address-cells = <1>;
 			#size-cells = <0>;
-			reg = <0 0x1100b000 0 0x100>;
 			interrupts = <GIC_SPI 141 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&topckgen CLK_TOP_MPLL_D2>,
 				 <&topckgen CLK_TOP_SPIM_MST_SEL>,
@@ -388,7 +388,6 @@
 		};
 
 		thermal: thermal@1100c800 {
-			#thermal-sensor-cells = <1>;
 			compatible = "mediatek,mt7986-thermal";
 			reg = <0 0x1100c800 0 0x800>;
 			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
@@ -396,30 +395,30 @@
 				 <&infracfg CLK_INFRA_ADC_26M_CK>,
 				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
 			clock-names = "therm", "auxadc", "adc_32k";
-			mediatek,auxadc = <&auxadc>;
-			mediatek,apmixedsys = <&apmixedsys>;
 			nvmem-cells = <&thermal_calibration>;
 			nvmem-cell-names = "calibration-data";
+			#thermal-sensor-cells = <1>;
+			mediatek,auxadc = <&auxadc>;
+			mediatek,apmixedsys = <&apmixedsys>;
 		};
 
 		pcie: pcie@11280000 {
 			compatible = "mediatek,mt7986-pcie",
 				     "mediatek,mt8192-pcie";
+			reg = <0x00 0x11280000 0x00 0x4000>;
+			reg-names = "pcie-mac";
+			ranges = <0x82000000 0x00 0x20000000 0x00
+				  0x20000000 0x00 0x10000000>;
 			device_type = "pci";
 			#address-cells = <3>;
 			#size-cells = <2>;
-			reg = <0x00 0x11280000 0x00 0x4000>;
-			reg-names = "pcie-mac";
 			interrupts = <GIC_SPI 168 IRQ_TYPE_LEVEL_HIGH>;
 			bus-range = <0x00 0xff>;
-			ranges = <0x82000000 0x00 0x20000000 0x00
-				  0x20000000 0x00 0x10000000>;
 			clocks = <&infracfg CLK_INFRA_IPCIE_PIPE_CK>,
 				 <&infracfg CLK_INFRA_IPCIE_CK>,
 				 <&infracfg CLK_INFRA_IPCIER_CK>,
 				 <&infracfg CLK_INFRA_IPCIEB_CK>;
 			clock-names = "pl_250m", "tl_26m", "peri_26m", "top_133m";
-			status = "disabled";
 
 			phys = <&pcie_port PHY_TYPE_PCIE>;
 			phy-names = "pcie-phy";
@@ -430,6 +429,8 @@
 					<0 0 0 2 &pcie_intc 1>,
 					<0 0 0 3 &pcie_intc 2>,
 					<0 0 0 4 &pcie_intc 3>;
+			status = "disabled";
+
 			pcie_intc: interrupt-controller {
 				#address-cells = <0>;
 				#interrupt-cells = <1>;
@@ -440,9 +441,9 @@
 		pcie_phy: t-phy {
 			compatible = "mediatek,mt7986-tphy",
 				     "mediatek,generic-tphy-v2";
+			ranges;
 			#address-cells = <2>;
 			#size-cells = <2>;
-			ranges;
 			status = "disabled";
 
 			pcie_port: pcie-phy@11c00000 {
@@ -467,9 +468,9 @@
 		usb_phy: t-phy@11e10000 {
 			compatible = "mediatek,mt7986-tphy",
 				     "mediatek,generic-tphy-v2";
+			ranges = <0 0 0x11e10000 0x1700>;
 			#address-cells = <1>;
 			#size-cells = <1>;
-			ranges = <0 0 0x11e10000 0x1700>;
 			status = "disabled";
 
 			u2port0: usb-phy@0 {
@@ -497,11 +498,11 @@
 		};
 
 		ethsys: syscon@15000000 {
-			 #address-cells = <1>;
-			 #size-cells = <1>;
 			 compatible = "mediatek,mt7986-ethsys",
 				      "syscon";
 			 reg = <0 0x15000000 0 0x1000>;
+			 #address-cells = <1>;
+			 #size-cells = <1>;
 			 #clock-cells = <1>;
 			 #reset-cells = <1>;
 		};
@@ -578,26 +579,26 @@
 					  <&topckgen CLK_TOP_SGM_325M_SEL>;
 			assigned-clock-parents = <&apmixedsys CLK_APMIXED_NET2PLL>,
 						 <&apmixedsys CLK_APMIXED_SGMPLL>;
+			#reset-cells = <1>;
+			#address-cells = <1>;
+			#size-cells = <0>;
 			mediatek,ethsys = <&ethsys>;
 			mediatek,sgmiisys = <&sgmiisys0>, <&sgmiisys1>;
 			mediatek,wed-pcie = <&wed_pcie>;
 			mediatek,wed = <&wed0>, <&wed1>;
-			#reset-cells = <1>;
-			#address-cells = <1>;
-			#size-cells = <0>;
 			status = "disabled";
 		};
 
 		wifi: wifi@18000000 {
 			compatible = "mediatek,mt7986-wmac";
+			reg = <0 0x18000000 0 0x1000000>,
+			      <0 0x10003000 0 0x1000>,
+			      <0 0x11d10000 0 0x1000>;
 			resets = <&watchdog MT7986_TOPRGU_CONSYS_SW_RST>;
 			reset-names = "consys";
 			clocks = <&topckgen CLK_TOP_CONN_MCUSYS_SEL>,
 				 <&topckgen CLK_TOP_AP2CNN_HOST_SEL>;
 			clock-names = "mcu", "ap2conn";
-			reg = <0 0x18000000 0 0x1000000>,
-			      <0 0x10003000 0 0x1000>,
-			      <0 0x11d10000 0 0x1000>;
 			interrupts = <GIC_SPI 213 IRQ_TYPE_LEVEL_HIGH>,
 				     <GIC_SPI 214 IRQ_TYPE_LEVEL_HIGH>,
 				     <GIC_SPI 215 IRQ_TYPE_LEVEL_HIGH>,
Signature · 1 line
2.43.0
author maintainer
#26 lore raw

PATCH 26/228 arm64: dts: mediatek: mt7986: drop invalid properties from ethsys

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3b449bfd2ff6c5d3ceecfcb18528ff8e1b4ac2fd ]

Mediatek ethsys controller / syscon binding doesn't allow any subnodes so "#address-cells" and "#size-cells" are redundant (actually:
disallowed).

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: syscon@15000000: '#address-cells', '#size-cells' do not match any of the regexes: 'pinctrl-[0-9]+'
from schema $id: http://devicetree.org/schemas/clock/mediatek,ethsys.yaml#

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 2 --
 1 file changed, 2 deletions(-)
arch/arm64/boot/dts/mediatek/mt7986a.dtsi +0 -2
index eba5e27a1bbea..850b664dfa13d 100644
@@ -501,8 +501,6 @@
 			 compatible = "mediatek,mt7986-ethsys",
 				      "syscon";
 			 reg = <0 0x15000000 0 0x1000>;
-			 #address-cells = <1>;
-			 #size-cells = <1>;
 			 #clock-cells = <1>;
 			 #reset-cells = <1>;
 		};
Signature · 1 line
2.43.0
author maintainer
#27 lore raw

PATCH 27/228 arm64: dts: mediatek: mt7986: drop "#reset-cells" from Ethernet controller

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 9bd88afc94c3570289a0f1c696578b3e1f4e3169 ]

Ethernet block doesn't include or act as a reset controller. Documentation also doesn't document "#reset-cells" for it.

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: ethernet@15100000: Unevaluated properties are not allowed ('#reset-cells' was unexpected)
from schema $id: http://devicetree.org/schemas/net/mediatek,net.yaml#

Fixes 082ff36bd5c0 ("arm64: dts: mediatek: mt7986: introduce ethernet nodes")
Cc Lorenzo Bianconi <[email protected]>
---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 1 -
 1 file changed, 1 deletion(-)
arch/arm64/boot/dts/mediatek/mt7986a.dtsi +0 -1
index 850b664dfa13d..5c2fe2f43a142 100644
@@ -577,7 +577,6 @@
 					  <&topckgen CLK_TOP_SGM_325M_SEL>;
 			assigned-clock-parents = <&apmixedsys CLK_APMIXED_NET2PLL>,
 						 <&apmixedsys CLK_APMIXED_SGMPLL>;
-			#reset-cells = <1>;
 			#address-cells = <1>;
 			#size-cells = <0>;
 			mediatek,ethsys = <&ethsys>;
Signature · 1 line
2.43.0
author maintainer
#28 lore raw

PATCH 28/228 arm64: dts: mediatek: mt7986: reorder nodes

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3f79e8f3364499750d7442767b101b7bc5864ddf ]

Use order described as preferred in DTS Coding Style:
1. Sort bus nodes by unit address
2. Use alpha-numerical order for the rest

Stable-dep-of: 970f8b01bd77 ("arm64: dts: mediatek: mt7986: drop invalid thermal block clock")

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 116 +++++++++++-----------
 1 file changed, 58 insertions(+), 58 deletions(-)
arch/arm64/boot/dts/mediatek/mt7986a.dtsi +58 -58
index 5c2fe2f43a142..f3a2a89fada41 100644
@@ -16,13 +16,6 @@
 	#address-cells = <2>;
 	#size-cells = <2>;
 
-	clk40m: oscillator-40m {
-		compatible = "fixed-clock";
-		clock-frequency = <40000000>;
-		#clock-cells = <0>;
-		clock-output-names = "clkxtal";
-	};
-
 	cpus {
 		#address-cells = <1>;
 		#size-cells = <0>;
@@ -59,6 +52,13 @@
 		};
 	};
 
+	clk40m: oscillator-40m {
+		compatible = "fixed-clock";
+		clock-frequency = <40000000>;
+		#clock-cells = <0>;
+		clock-output-names = "clkxtal";
+	};
+
 	psci {
 		compatible = "arm,psci-0.2";
 		method = "smc";
@@ -121,15 +121,6 @@
 
 	};
 
-	timer {
-		compatible = "arm,armv8-timer";
-		interrupt-parent = <&gic>;
-		interrupts = <GIC_PPI 13 IRQ_TYPE_LEVEL_LOW>,
-			     <GIC_PPI 14 IRQ_TYPE_LEVEL_LOW>,
-			     <GIC_PPI 11 IRQ_TYPE_LEVEL_LOW>,
-			     <GIC_PPI 10 IRQ_TYPE_LEVEL_LOW>;
-	};
-
 	soc {
 		compatible = "simple-bus";
 		ranges;
@@ -203,6 +194,19 @@
 			#interrupt-cells = <2>;
 		};
 
+		pwm: pwm@10048000 {
+			compatible = "mediatek,mt7986-pwm";
+			reg = <0 0x10048000 0 0x1000>;
+			#pwm-cells = <2>;
+			interrupts = <GIC_SPI 137 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&topckgen CLK_TOP_PWM_SEL>,
+				 <&infracfg CLK_INFRA_PWM_STA>,
+				 <&infracfg CLK_INFRA_PWM1_CK>,
+				 <&infracfg CLK_INFRA_PWM2_CK>;
+			clock-names = "top", "main", "pwm1", "pwm2";
+			status = "disabled";
+		};
+
 		sgmiisys0: syscon@10060000 {
 			compatible = "mediatek,mt7986-sgmiisys_0",
 				     "syscon";
@@ -240,19 +244,6 @@
 			status = "disabled";
 		};
 
-		pwm: pwm@10048000 {
-			compatible = "mediatek,mt7986-pwm";
-			reg = <0 0x10048000 0 0x1000>;
-			#pwm-cells = <2>;
-			interrupts = <GIC_SPI 137 IRQ_TYPE_LEVEL_HIGH>;
-			clocks = <&topckgen CLK_TOP_PWM_SEL>,
-				 <&infracfg CLK_INFRA_PWM_STA>,
-				 <&infracfg CLK_INFRA_PWM1_CK>,
-				 <&infracfg CLK_INFRA_PWM2_CK>;
-			clock-names = "top", "main", "pwm1", "pwm2";
-			status = "disabled";
-		};
-
 		uart0: serial@11002000 {
 			compatible = "mediatek,mt7986-uart",
 				     "mediatek,mt6577-uart";
@@ -336,6 +327,21 @@
 			status = "disabled";
 		};
 
+		thermal: thermal@1100c800 {
+			compatible = "mediatek,mt7986-thermal";
+			reg = <0 0x1100c800 0 0x800>;
+			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&infracfg CLK_INFRA_THERM_CK>,
+				 <&infracfg CLK_INFRA_ADC_26M_CK>,
+				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
+			clock-names = "therm", "auxadc", "adc_32k";
+			nvmem-cells = <&thermal_calibration>;
+			nvmem-cell-names = "calibration-data";
+			#thermal-sensor-cells = <1>;
+			mediatek,auxadc = <&auxadc>;
+			mediatek,apmixedsys = <&apmixedsys>;
+		};
+
 		auxadc: adc@1100d000 {
 			compatible = "mediatek,mt7986-auxadc";
 			reg = <0 0x1100d000 0 0x1000>;
@@ -387,21 +393,6 @@
 			status = "disabled";
 		};
 
-		thermal: thermal@1100c800 {
-			compatible = "mediatek,mt7986-thermal";
-			reg = <0 0x1100c800 0 0x800>;
-			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
-			clocks = <&infracfg CLK_INFRA_THERM_CK>,
-				 <&infracfg CLK_INFRA_ADC_26M_CK>,
-				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
-			clock-names = "therm", "auxadc", "adc_32k";
-			nvmem-cells = <&thermal_calibration>;
-			nvmem-cell-names = "calibration-data";
-			#thermal-sensor-cells = <1>;
-			mediatek,auxadc = <&auxadc>;
-			mediatek,apmixedsys = <&apmixedsys>;
-		};
-
 		pcie: pcie@11280000 {
 			compatible = "mediatek,mt7986-pcie",
 				     "mediatek,mt8192-pcie";
@@ -531,20 +522,6 @@
 			mediatek,wo-ccif = <&wo_ccif1>;
 		};
 
-		wo_ccif0: syscon@151a5000 {
-			compatible = "mediatek,mt7986-wo-ccif", "syscon";
-			reg = <0 0x151a5000 0 0x1000>;
-			interrupt-parent = <&gic>;
-			interrupts = <GIC_SPI 211 IRQ_TYPE_LEVEL_HIGH>;
-		};
-
-		wo_ccif1: syscon@151ad000 {
-			compatible = "mediatek,mt7986-wo-ccif", "syscon";
-			reg = <0 0x151ad000 0 0x1000>;
-			interrupt-parent = <&gic>;
-			interrupts = <GIC_SPI 212 IRQ_TYPE_LEVEL_HIGH>;
-		};
-
 		eth: ethernet@15100000 {
 			compatible = "mediatek,mt7986-eth";
 			reg = <0 0x15100000 0 0x80000>;
@@ -586,6 +563,20 @@
 			status = "disabled";
 		};
 
+		wo_ccif0: syscon@151a5000 {
+			compatible = "mediatek,mt7986-wo-ccif", "syscon";
+			reg = <0 0x151a5000 0 0x1000>;
+			interrupt-parent = <&gic>;
+			interrupts = <GIC_SPI 211 IRQ_TYPE_LEVEL_HIGH>;
+		};
+
+		wo_ccif1: syscon@151ad000 {
+			compatible = "mediatek,mt7986-wo-ccif", "syscon";
+			reg = <0 0x151ad000 0 0x1000>;
+			interrupt-parent = <&gic>;
+			interrupts = <GIC_SPI 212 IRQ_TYPE_LEVEL_HIGH>;
+		};
+
 		wifi: wifi@18000000 {
 			compatible = "mediatek,mt7986-wmac";
 			reg = <0 0x18000000 0 0x1000000>,
@@ -643,4 +634,13 @@
 			};
 		};
 	};
+
+	timer {
+		compatible = "arm,armv8-timer";
+		interrupt-parent = <&gic>;
+		interrupts = <GIC_PPI 13 IRQ_TYPE_LEVEL_LOW>,
+			     <GIC_PPI 14 IRQ_TYPE_LEVEL_LOW>,
+			     <GIC_PPI 11 IRQ_TYPE_LEVEL_LOW>,
+			     <GIC_PPI 10 IRQ_TYPE_LEVEL_LOW>;
+	};
 };
Signature · 1 line
2.43.0
author maintainer
#29 lore raw

PATCH 29/228 arm64: dts: mediatek: mt7986: drop invalid thermal block clock

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 970f8b01bd7719a22e577ba6c78e27f9ccf22783 ]

Thermal block uses only two clocks. Its binding doesn't document or allow "adc_32k". Also Linux driver doesn't support it.

It has been additionally verified by Angelo by his detailed research on MT7981 / MT7986 clocks (thanks!).

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: thermal@1100c800: clocks: [[4, 27], [4, 44], [4, 45]] is too long
from schema $id: http://devicetree.org/schemas/thermal/mediatek,thermal.yaml#
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: thermal@1100c800: clock-names: ['therm', 'auxadc', 'adc_32k'] is too long
from schema $id: http://devicetree.org/schemas/thermal/mediatek,thermal.yaml#

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/mediatek/mt7986a.dtsi +2 -3
index f3a2a89fada41..559990dcd1d17 100644
@@ -332,9 +332,8 @@
 			reg = <0 0x1100c800 0 0x800>;
 			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&infracfg CLK_INFRA_THERM_CK>,
-				 <&infracfg CLK_INFRA_ADC_26M_CK>,
-				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
-			clock-names = "therm", "auxadc", "adc_32k";
+				 <&infracfg CLK_INFRA_ADC_26M_CK>;
+			clock-names = "therm", "auxadc";
 			nvmem-cells = <&thermal_calibration>;
 			nvmem-cell-names = "calibration-data";
 			#thermal-sensor-cells = <1>;
Signature · 1 line
2.43.0
author maintainer
#30 lore raw

PATCH 30/228 arm64: dts: mediatek: mt7986: prefix BPI-R3 cooling maps with "map-"

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit f8c65a5e4560781f2ea175d8f26cd75ac98e8d78 ]

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: thermal-zones: cpu-thermal:cooling-maps: 'cpu-active-high', 'cpu-active-low', 'cpu-active-med' do not match any of the regexes: '^map[-a-zA-Z0-9]*$', 'pinctrl-[0-9]+'
from schema $id: http://devicetree.org/schemas/thermal/thermal-zones.yaml#

Fixes c26f779a2295 ("arm64: dts: mt7986: add pwm-fan and cooling-maps to BPI-R3 dts")
Cc Daniel Golle <[email protected]>
---
 arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dts | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dts +3 -3
index e04b1c0c0ebbf..ed79ad1ae8716 100644
@@ -146,19 +146,19 @@
 
 &cpu_thermal {
 	cooling-maps {
-		cpu-active-high {
+		map-cpu-active-high {
 			/* active: set fan to cooling level 2 */
 			cooling-device = <&fan 2 2>;
 			trip = <&cpu_trip_active_high>;
 		};
 
-		cpu-active-med {
+		map-cpu-active-med {
 			/* active: set fan to cooling level 1 */
 			cooling-device = <&fan 1 1>;
 			trip = <&cpu_trip_active_med>;
 		};
 
-		cpu-active-low {
+		map-cpu-active-low {
 			/* active: set fan to cooling level 0 */
 			cooling-device = <&fan 0 0>;
 			trip = <&cpu_trip_active_low>;
Signature · 1 line
2.43.0
author maintainer
#31 lore raw

PATCH 31/228 arm64: dts: mediatek: mt2712: fix validation errors

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3baac7291effb501c4d52df7019ebf52011e5772 ]

1. Fixup infracfg clock controller binding
It also acts as reset controller so #reset-cells is required.
2. Use -pins suffix for pinctrl

This fixes:
arch/arm64/boot/dts/mediatek/mt2712-evb.dtb: syscon@10001000: '#reset-cells' is a required property
from schema $id: http://devicetree.org/schemas/arm/mediatek/mediatek,infracfg.yaml#
arch/arm64/boot/dts/mediatek/mt2712-evb.dtb: pinctrl@1000b000: 'eth_default', 'eth_sleep', 'usb0_iddig', 'usb1_iddig' do not match any of the regexes: 'pinctrl-[0-9]+', 'pins$'
from schema $id: http://devicetree.org/schemas/pinctrl/mediatek,mt65xx-pinctrl.yaml#

[Angelo: Added Fixes tags]
Fixes 5d4839709c8e ("arm64: dts: mt2712: Add clock controller device nodes")
Fixes 1724f4cc5133 ("arm64: dts: Add USB3 related nodes for MT2712")
---
 arch/arm64/boot/dts/mediatek/mt2712-evb.dts | 8 ++++----
 arch/arm64/boot/dts/mediatek/mt2712e.dtsi   | 3 ++-
 2 files changed, 6 insertions(+), 5 deletions(-)
arch/arm64/boot/dts/mediatek/mt2712-evb.dts +4 -4
index fffdb7bbf889e..2d0ef6f23b3a9 100644
@@ -129,7 +129,7 @@
 };
 
 &pio {
-	eth_default: eth_default {
+	eth_default: eth-default-pins {
 		tx_pins {
 			pinmux = <MT2712_PIN_71_GBE_TXD3__FUNC_GBE_TXD3>,
 				 <MT2712_PIN_72_GBE_TXD2__FUNC_GBE_TXD2>,
@@ -156,7 +156,7 @@
 		};
 	};
 
-	eth_sleep: eth_sleep {
+	eth_sleep: eth-sleep-pins {
 		tx_pins {
 			pinmux = <MT2712_PIN_71_GBE_TXD3__FUNC_GPIO71>,
 				 <MT2712_PIN_72_GBE_TXD2__FUNC_GPIO72>,
@@ -182,14 +182,14 @@
 		};
 	};
 
-	usb0_id_pins_float: usb0_iddig {
+	usb0_id_pins_float: usb0-iddig-pins {
 		pins_iddig {
 			pinmux = <MT2712_PIN_12_IDDIG_P0__FUNC_IDDIG_A>;
 			bias-pull-up;
 		};
 	};
 
-	usb1_id_pins_float: usb1_iddig {
+	usb1_id_pins_float: usb1-iddig-pins {
 		pins_iddig {
 			pinmux = <MT2712_PIN_14_IDDIG_P1__FUNC_IDDIG_B>;
 			bias-pull-up;
arch/arm64/boot/dts/mediatek/mt2712e.dtsi +2 -1
index ed1a9d3194153..f767f921bdee1 100644
@@ -249,10 +249,11 @@
 		#clock-cells = <1>;
 	};
 
-	infracfg: syscon@10001000 {
+	infracfg: clock-controller@10001000 {
 		compatible = "mediatek,mt2712-infracfg", "syscon";
 		reg = <0 0x10001000 0 0x1000>;
 		#clock-cells = <1>;
+		#reset-cells = <1>;
 	};
 
 	pericfg: syscon@10003000 {
Signature · 1 line
2.43.0
author maintainer
#32 lore raw

PATCH 32/228 arm64: dts: rockchip: mark system power controller and fix typo on orangepi-5-plus

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Muhammed Efe Cetin <[email protected]>

[ Upstream commit 08cd20bdecd9cfde5c1aec6146fa22ca753efea1 ]

Mark the PMIC as system power controller, so the board will shut-down properly and fix the typo on rk806_dvs1_null pins property.

Fixes 236d225e1ee7 ("arm64: dts: rockchip: Add board device tree for rk3588-orangepi-5-plus")
---
 arch/arm64/boot/dts/rockchip/rk3588-orangepi-5-plus.dts | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/arm64/boot/dts/rockchip/rk3588-orangepi-5-plus.dts +2 -1
index 3e660ff6cd5ff..e74871491ef56 100644
@@ -486,6 +486,7 @@
 		pinctrl-0 = <&pmic_pins>, <&rk806_dvs1_null>,
 			    <&rk806_dvs2_null>, <&rk806_dvs3_null>;
 		spi-max-frequency = <1000000>;
+		system-power-controller;
 
 		vcc1-supply = <&vcc5v0_sys>;
 		vcc2-supply = <&vcc5v0_sys>;
@@ -507,7 +508,7 @@
 		#gpio-cells = <2>;
 
 		rk806_dvs1_null: dvs1-null-pins {
-			pins = "gpio_pwrctrl2";
+			pins = "gpio_pwrctrl1";
 			function = "pin_fun0";
 		};
 
Signature · 1 line
2.43.0
author maintainer
#33 lore raw

PATCH 33/228 arm64: dts: rockchip: regulator for sd needs to be always on for BPI-R2Pro

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jose Ignacio Tornos Martinez <[email protected]>

[ Upstream commit 433d54818f64a2fe0562f8c04c7a81f562368515 ]

With default dts configuration for BPI-R2Pro, the regulator for sd card is powered off when reboot is commanded, and the only solution to detect the sd card again, and therefore, allow rebooting from there, is to do a hardware reset.

Configure the regulator for sd to be always on for BPI-R2Pro in order to avoid this issue.

---
 arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts | 2 ++
 1 file changed, 2 insertions(+)
arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts +2 -0
index 87c45d8be420f..dc5892d25c100 100644
@@ -416,6 +416,8 @@
 
 			vccio_sd: LDO_REG5 {
 				regulator-name = "vccio_sd";
+				regulator-always-on;
+				regulator-boot-on;
 				regulator-min-microvolt = <1800000>;
 				regulator-max-microvolt = <3300000>;
 
Signature · 1 line
2.43.0
author maintainer
#34 lore raw

PATCH 34/228 block: fix module reference leakage from bdev_open_by_dev error path

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yu Kuai <[email protected]>

[ Upstream commit 9617cd6f24b294552a817f80f5225431ef67b540 ]

At the time bdev_may_open() is called, module reference is grabbed already, hence module reference should be released if bdev_may_open() failed.

This problem is found by code review.

Fixes ed5cc702d311 ("block: Add config option to not allow writing to mounted devices")
---
 block/bdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
block/bdev.c +1 -1
index 2b0f97651a0a7..812872cdc0bef 100644
@@ -873,7 +873,7 @@ struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 		goto abort_claiming;
 	ret = -EBUSY;
 	if (!bdev_may_open(bdev, mode))
-		goto abort_claiming;
+		goto put_module;
 	if (bdev_is_partition(bdev))
 		ret = blkdev_get_part(bdev, mode);
 	else
Signature · 1 line
2.43.0
author maintainer
#35 lore raw

PATCH 35/228 arm64: dts: qcom: Fix type of "wdog" IRQs for remoteprocs

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luca Weiss <[email protected]>

[ Upstream commit f011688162ec4c492c12ee7cced74c097270baa2 ]

The code in qcom_q6v5_init() requests the "wdog" IRQ as IRQF_TRIGGER_RISING. If dt defines the interrupt type as LEVEL_HIGH then the driver will have issues getting the IRQ again after probe deferral with an error like:

irq: type mismatch, failed to map hwirq-14 for interrupt-controller@b220000!

Fix that by updating the devicetrees to use IRQ_TYPE_EDGE_RISING for these interrupts, as is already used in most dt's. Also the driver was already using the interrupts with that type.

Fixes 3658e411efcb ("arm64: dts: qcom: sc7280: Add ADSP node")
Fixes df62402e5ff9 ("arm64: dts: qcom: sc7280: Add CDSP node")
Fixes 152d1faf1e2f ("arm64: dts: qcom: add SC8280XP platform")
Fixes 8eb5287e8a42 ("arm64: dts: qcom: sm6350: Add CDSP nodes")
Fixes efc33c969f23 ("arm64: dts: qcom: sm6350: Add ADSP nodes")
Fixes fe6fd26aeddf ("arm64: dts: qcom: sm6375: Add ADSP&CDSP")
Fixes 23a8903785b9 ("arm64: dts: qcom: sm8250: Add remoteprocs")
[bjorn: Added fixes references]
---
 arch/arm64/boot/dts/qcom/sc7280.dtsi   | 4 ++--
 arch/arm64/boot/dts/qcom/sc8280xp.dtsi | 6 +++---
 arch/arm64/boot/dts/qcom/sm6350.dtsi   | 4 ++--
 arch/arm64/boot/dts/qcom/sm6375.dtsi   | 2 +-
 arch/arm64/boot/dts/qcom/sm8250.dtsi   | 6 +++---
 5 files changed, 11 insertions(+), 11 deletions(-)
arch/arm64/boot/dts/qcom/sc7280.dtsi +2 -2
index 7dc2c37716e84..46545cd6964a9 100644
@@ -3650,7 +3650,7 @@
 			compatible = "qcom,sc7280-adsp-pas";
 			reg = <0 0x03700000 0 0x100>;
 
-			interrupts-extended = <&pdc 6 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&pdc 6 IRQ_TYPE_EDGE_RISING>,
 					      <&adsp_smp2p_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&adsp_smp2p_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&adsp_smp2p_in 2 IRQ_TYPE_EDGE_RISING>,
@@ -3887,7 +3887,7 @@
 			compatible = "qcom,sc7280-cdsp-pas";
 			reg = <0 0x0a300000 0 0x10000>;
 
-			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_EDGE_RISING>,
 					      <&cdsp_smp2p_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&cdsp_smp2p_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&cdsp_smp2p_in 2 IRQ_TYPE_EDGE_RISING>,
arch/arm64/boot/dts/qcom/sc8280xp.dtsi +3 -3
index febf28356ff8b..e47c58e06c395 100644
@@ -2635,7 +2635,7 @@
 			compatible = "qcom,sc8280xp-adsp-pas";
 			reg = <0 0x03000000 0 0x100>;
 
-			interrupts-extended = <&intc GIC_SPI 162 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&intc GIC_SPI 162 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 2 IRQ_TYPE_EDGE_RISING>,
@@ -4407,7 +4407,7 @@
 			compatible = "qcom,sc8280xp-nsp0-pas";
 			reg = <0 0x1b300000 0 0x100>;
 
-			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_nsp0_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_nsp0_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_nsp0_in 2 IRQ_TYPE_EDGE_RISING>,
@@ -4538,7 +4538,7 @@
 			compatible = "qcom,sc8280xp-nsp1-pas";
 			reg = <0 0x21300000 0 0x100>;
 
-			interrupts-extended = <&intc GIC_SPI 887 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&intc GIC_SPI 887 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_nsp1_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_nsp1_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_nsp1_in 2 IRQ_TYPE_EDGE_RISING>,
arch/arm64/boot/dts/qcom/sm6350.dtsi +2 -2
index 43cffe8e1247e..7532107580aee 100644
@@ -1249,7 +1249,7 @@
 			compatible = "qcom,sm6350-adsp-pas";
 			reg = <0 0x03000000 0 0x100>;
 
-			interrupts-extended = <&pdc 6 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&pdc 6 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 2 IRQ_TYPE_EDGE_RISING>,
@@ -1509,7 +1509,7 @@
 			compatible = "qcom,sm6350-cdsp-pas";
 			reg = <0 0x08300000 0 0x10000>;
 
-			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_cdsp_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_cdsp_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_cdsp_in 2 IRQ_TYPE_EDGE_RISING>,
arch/arm64/boot/dts/qcom/sm6375.dtsi +1 -1
index 7ac8bf26dda3a..a74472ac6186d 100644
@@ -1559,7 +1559,7 @@
 			compatible = "qcom,sm6375-adsp-pas";
 			reg = <0 0x0a400000 0 0x100>;
 
-			interrupts-extended = <&intc GIC_SPI 282 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&intc GIC_SPI 282 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 2 IRQ_TYPE_EDGE_RISING>,
arch/arm64/boot/dts/qcom/sm8250.dtsi +3 -3
index 760501c1301a6..b07a3bad6eb96 100644
@@ -3025,7 +3025,7 @@
 			compatible = "qcom,sm8250-slpi-pas";
 			reg = <0 0x05c00000 0 0x4000>;
 
-			interrupts-extended = <&pdc 9 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&pdc 9 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_slpi_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_slpi_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_slpi_in 2 IRQ_TYPE_EDGE_RISING>,
@@ -3729,7 +3729,7 @@
 			compatible = "qcom,sm8250-cdsp-pas";
 			reg = <0 0x08300000 0 0x10000>;
 
-			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&intc GIC_SPI 578 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_cdsp_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_cdsp_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_cdsp_in 2 IRQ_TYPE_EDGE_RISING>,
@@ -5887,7 +5887,7 @@
 			compatible = "qcom,sm8250-adsp-pas";
 			reg = <0 0x17300000 0 0x100>;
 
-			interrupts-extended = <&pdc 6 IRQ_TYPE_LEVEL_HIGH>,
+			interrupts-extended = <&pdc 6 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 0 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 1 IRQ_TYPE_EDGE_RISING>,
 					      <&smp2p_adsp_in 2 IRQ_TYPE_EDGE_RISING>,
Signature · 1 line
2.43.0
author maintainer
#36 lore raw

PATCH 36/228 arm64: dts: qcom: x1e80100: Fix the compatible for cluster idle states

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rajendra Nayak <[email protected]>

[ Upstream commit cb939b9b35426852896790aba2f18f46df34e596 ]

The compatible's for the cluster/domain idle states of x1e80100 are wrong, fix it.

Fixes af16b00578a7 ("arm64: dts: qcom: Add base X1E80100 dtsi and the QCP dts")
---
 arch/arm64/boot/dts/qcom/x1e80100.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/qcom/x1e80100.dtsi +2 -2
index be1285d9919e0..5ba7924aa4355 100644
@@ -278,7 +278,7 @@
 
 		domain-idle-states {
 			CLUSTER_CL4: cluster-sleep-0 {
-				compatible = "arm,idle-state";
+				compatible = "domain-idle-state";
 				idle-state-name = "l2-ret";
 				arm,psci-suspend-param = <0x01000044>;
 				entry-latency-us = <350>;
@@ -287,7 +287,7 @@
 			};
 
 			CLUSTER_CL5: cluster-sleep-1 {
-				compatible = "arm,idle-state";
+				compatible = "domain-idle-state";
 				idle-state-name = "ret-pll-off";
 				arm,psci-suspend-param = <0x01000054>;
 				entry-latency-us = <2200>;
Signature · 1 line
2.43.0
author maintainer
#37 lore raw

PATCH 37/228 arm64: dts: qcom: sc8180x: Fix ss_phy_irq for secondary USB controller

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maximilian Luz <[email protected]>

[ Upstream commit ecda8309098402f878c96184f29a1b7ec682d772 ]

The ACPI DSDT of the Surface Pro X (SQ2) specifies the interrupts for the secondary UBS controller as

    Name (_CRS, ResourceTemplate ()
    {
        Interrupt (ResourceConsumer, Level, ActiveHigh, Shared, ,, )
        {
            0x000000AA,
        }
        Interrupt (ResourceConsumer, Level, ActiveHigh, SharedAndWake, ,, )
        {
            0x000000A7,     // hs_phy_irq: &intc GIC_SPI 136
        }
        Interrupt (ResourceConsumer, Level, ActiveHigh, SharedAndWake, ,, )
        {
            0x00000228,     // ss_phy_irq: &pdc 40
        }
        Interrupt (ResourceConsumer, Edge, ActiveHigh, SharedAndWake, ,, )
        {
            0x0000020A,     // dm_hs_phy_irq: &pdc 10
        }
        Interrupt (ResourceConsumer, Edge, ActiveHigh, SharedAndWake, ,, )
        {
            0x0000020B,     // dp_hs_phy_irq: &pdc 11
        }
    })

Generally, the interrupts above 0x200 map to the PDC interrupts (as used in the devicetree) as ACPI_NUMBER - 0x200. Note that this lines up with dm_hs_phy_irq and dp_hs_phy_irq (as well as the interrupts for the primary USB controller).

Based on the snippet above, ss_phy_irq should therefore be PDC 40 (= 0x28) and not PDC 7. The latter is according to ACPI instead used as ss_phy_irq for port 0 of the multiport USB controller). Fix this by setting ss_phy_irq to '&pdc 40'.

Fixes b080f53a8f44 ("arm64: dts: qcom: sc8180x: Add remoteprocs, wifi and usb nodes")
---
 arch/arm64/boot/dts/qcom/sc8180x.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/qcom/sc8180x.dtsi +1 -1
index c0dd44f146748..b481c15b255ce 100644
@@ -2641,7 +2641,7 @@
 			resets = <&gcc GCC_USB30_SEC_BCR>;
 			power-domains = <&gcc USB30_SEC_GDSC>;
 			interrupts-extended = <&intc GIC_SPI 136 IRQ_TYPE_LEVEL_HIGH>,
-					      <&pdc 7 IRQ_TYPE_LEVEL_HIGH>,
+					      <&pdc 40 IRQ_TYPE_LEVEL_HIGH>,
 					      <&pdc 10 IRQ_TYPE_EDGE_BOTH>,
 					      <&pdc 11 IRQ_TYPE_EDGE_BOTH>;
 			interrupt-names = "hs_phy_irq", "ss_phy_irq",
Signature · 1 line
2.43.0
author maintainer
#38 lore raw

PATCH 38/228 gpio: tangier: Use correct type for the IRQ chip data

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

[ Upstream commit 7d045025a24b6336d444d359bd4312f351d017f9 ]

IRQ chip data contains a pointer to the GPIO chip. Luckily we have the pointers the same, but strictly speaking it's not guaranteed. Even though, still better to fix this.

Fixes ccf6fd6dcc86 ("gpio: merrifield: Introduce GPIO driver to support Merrifield")
---
 drivers/gpio/gpio-tangier.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/gpio/gpio-tangier.c +6 -3
index b75e0b12087ac..4b29abafecf6a 100644
@@ -195,7 +195,8 @@ static int tng_gpio_set_config(struct gpio_chip *chip, unsigned int offset,
 
 static void tng_irq_ack(struct irq_data *d)
 {
-	struct tng_gpio *priv = irq_data_get_irq_chip_data(d);
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct tng_gpio *priv = gpiochip_get_data(gc);
 	irq_hw_number_t gpio = irqd_to_hwirq(d);
 	void __iomem *gisr;
 	u8 shift;
@@ -227,7 +228,8 @@ static void tng_irq_unmask_mask(struct tng_gpio *priv, u32 gpio, bool unmask)
 
 static void tng_irq_mask(struct irq_data *d)
 {
-	struct tng_gpio *priv = irq_data_get_irq_chip_data(d);
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct tng_gpio *priv = gpiochip_get_data(gc);
 	irq_hw_number_t gpio = irqd_to_hwirq(d);
 
 	tng_irq_unmask_mask(priv, gpio, false);
@@ -236,7 +238,8 @@ static void tng_irq_mask(struct irq_data *d)
 
 static void tng_irq_unmask(struct irq_data *d)
 {
-	struct tng_gpio *priv = irq_data_get_irq_chip_data(d);
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct tng_gpio *priv = gpiochip_get_data(gc);
 	irq_hw_number_t gpio = irqd_to_hwirq(d);
 
 	gpiochip_enable_irq(&priv->chip, gpio);
Signature · 1 line
2.43.0
author maintainer
#39 lore raw

PATCH 39/228 ARC: [plat-hsdk]: Remove misplaced interrupt-cells property

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexey Brodkin <[email protected]>

[ Upstream commit 61231eb8113ce47991f35024f9c20810b37996bf ]

"gmac" node stands for just an ordinary Ethernet controller, which is by no means a provider of interrupts, i.e. it doesn't serve as an interrupt controller, thus "#interrupt-cells" property doesn't belong to it and so we remove it.

Fixes:
------------>8------------
DTC arch/arc/boot/dts/hsdk.dtb
arch/arc/boot/dts/hsdk.dts:207.23-235.5: Warning (interrupt_provider): /soc/ethernet@8000: '#interrupt-cells' found, but node is not an interrupt provider arch/arc/boot/dts/hsdk.dtb: Warning (interrupt_map): Failed prerequisite 'interrupt_provider'
------------>8------------

---
 arch/arc/boot/dts/hsdk.dts | 1 -
 1 file changed, 1 deletion(-)
arch/arc/boot/dts/hsdk.dts +0 -1
index 6691f42550778..41b980df862b1 100644
@@ -205,7 +205,6 @@
 		};
 
 		gmac: ethernet@8000 {
-			#interrupt-cells = <1>;
 			compatible = "snps,dwmac";
 			reg = <0x8000 0x2000>;
 			interrupts = <10>;
Signature · 1 line
2.43.0
author maintainer
#40 lore raw

PATCH 40/228 wifi: mac80211: clean up assignments to pointer cache.

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Colin Ian King <[email protected]>

[ Upstream commit ba4b1fa3128b2fbf14e167230315cbd9074b629b ]

The assignment to pointer cache in function mesh_fast_tx_gc can be made at the declaration time rather than a later assignment. There are also 3 functions where pointer cache is being initialized at declaration time and later re-assigned again with the same value, these are redundant and can be removed.

Cleans up code and three clang scan build warnings:
warning: Value stored to 'cache' during its initialization is never read [deadcode.DeadStores]

Stable-dep-of: 8c75cdcdf869 ("wifi: mac80211: split mesh fast tx cache into local/proxied/forwarded")

---
 net/mac80211/mesh_pathtbl.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
net/mac80211/mesh_pathtbl.c +1 -5
index 735edde1bd819..91b55d6a68b97 100644
@@ -600,11 +600,10 @@ void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
 void mesh_fast_tx_gc(struct ieee80211_sub_if_data *sdata)
 {
 	unsigned long timeout = msecs_to_jiffies(MESH_FAST_TX_CACHE_TIMEOUT);
-	struct mesh_tx_cache *cache;
+	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
 	struct ieee80211_mesh_fast_tx *entry;
 	struct hlist_node *n;
 
-	cache = &sdata->u.mesh.tx_cache;
 	if (atomic_read(&cache->rht.nelems) < MESH_FAST_TX_CACHE_THRESHOLD_SIZE)
 		return;
 
@@ -622,7 +621,6 @@ void mesh_fast_tx_flush_mpath(struct mesh_path *mpath)
 	struct ieee80211_mesh_fast_tx *entry;
 	struct hlist_node *n;
 
-	cache = &sdata->u.mesh.tx_cache;
 	spin_lock_bh(&cache->walk_lock);
 	hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list)
 		if (entry->mpath == mpath)
@@ -637,7 +635,6 @@ void mesh_fast_tx_flush_sta(struct ieee80211_sub_if_data *sdata,
 	struct ieee80211_mesh_fast_tx *entry;
 	struct hlist_node *n;
 
-	cache = &sdata->u.mesh.tx_cache;
 	spin_lock_bh(&cache->walk_lock);
 	hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list)
 		if (rcu_access_pointer(entry->mpath->next_hop) == sta)
@@ -651,7 +648,6 @@ void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata,
 	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
 	struct ieee80211_mesh_fast_tx *entry;
 
-	cache = &sdata->u.mesh.tx_cache;
 	spin_lock_bh(&cache->walk_lock);
 	entry = rhashtable_lookup_fast(&cache->rht, addr, fast_tx_rht_params);
 	if (entry)
Signature · 1 line
2.43.0
author maintainer
#41 lore raw

PATCH 41/228 wifi: mac80211: split mesh fast tx cache into local/proxied/forwarded

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Felix Fietkau <[email protected]>

[ Upstream commit 8c75cdcdf869acabfdc7858827099dcde9f24e6c ]

Depending on the origin of the packets (and their SA), 802.11 + mesh headers could be filled in differently. In order to properly deal with that, add a new field to the lookup key, indicating the type (local, proxied or forwarded). This can fix spurious packet drop issues that depend on the order in which nodes/hosts communicate with each other.

Fixes d5edb9ae8d56 ("wifi: mac80211: mesh fast xmit support")
[use sizeof_field() for key_len]
---
 net/mac80211/mesh.c         |  8 +++++++-
 net/mac80211/mesh.h         | 36 +++++++++++++++++++++++++++++++++---
 net/mac80211/mesh_pathtbl.c | 31 ++++++++++++++++++++++---------
 net/mac80211/rx.c           | 13 ++++++++++---
 4 files changed, 72 insertions(+), 16 deletions(-)
net/mac80211/mesh.c +7 -1
index fccbcde3359af..3d4806b7ff6c4 100644
@@ -769,6 +769,9 @@ bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
 			      struct sk_buff *skb, u32 ctrl_flags)
 {
 	struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh;
+	struct ieee80211_mesh_fast_tx_key key = {
+		.type = MESH_FAST_TX_TYPE_LOCAL
+	};
 	struct ieee80211_mesh_fast_tx *entry;
 	struct ieee80211s_hdr *meshhdr;
 	u8 sa[ETH_ALEN] __aligned(2);
@@ -804,7 +807,10 @@ bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
 			return false;
 	}
 
-	entry = mesh_fast_tx_get(sdata, skb->data);
+	ether_addr_copy(key.addr, skb->data);
+	if (!ether_addr_equal(skb->data + ETH_ALEN, sdata->vif.addr))
+		key.type = MESH_FAST_TX_TYPE_PROXIED;
+	entry = mesh_fast_tx_get(sdata, &key);
 	if (!entry)
 		return false;
 
net/mac80211/mesh.h +33 -3
index ad8469293d712..58c619874ca6a 100644
@@ -133,10 +133,39 @@ struct mesh_path {
 #define MESH_FAST_TX_CACHE_THRESHOLD_SIZE	384
 #define MESH_FAST_TX_CACHE_TIMEOUT		8000 /* msecs */
 
+/**
+ * enum ieee80211_mesh_fast_tx_type - cached mesh fast tx entry type
+ *
+ * @MESH_FAST_TX_TYPE_LOCAL: tx from the local vif address as SA
+ * @MESH_FAST_TX_TYPE_PROXIED: local tx with a different SA (e.g. bridged)
+ * @MESH_FAST_TX_TYPE_FORWARDED: forwarded from a different mesh point
+ * @NUM_MESH_FAST_TX_TYPE: number of entry types
+ */
+enum ieee80211_mesh_fast_tx_type {
+	MESH_FAST_TX_TYPE_LOCAL,
+	MESH_FAST_TX_TYPE_PROXIED,
+	MESH_FAST_TX_TYPE_FORWARDED,
+
+	/* must be last */
+	NUM_MESH_FAST_TX_TYPE
+};
+
+
+/**
+ * struct ieee80211_mesh_fast_tx_key - cached mesh fast tx entry key
+ *
+ * @addr: The Ethernet DA for this entry
+ * @type: cache entry type
+ */
+struct ieee80211_mesh_fast_tx_key {
+	u8 addr[ETH_ALEN] __aligned(2);
+	u16 type;
+};
+
 /**
  * struct ieee80211_mesh_fast_tx - cached mesh fast tx entry
  * @rhash: rhashtable pointer
- * @addr_key: The Ethernet DA which is the key for this entry
+ * @key: the lookup key for this cache entry
  * @fast_tx: base fast_tx data
  * @hdr: cached mesh and rfc1042 headers
  * @hdrlen: length of mesh + rfc1042
@@ -147,7 +176,7 @@ struct mesh_path {
  */
 struct ieee80211_mesh_fast_tx {
 	struct rhash_head rhash;
-	u8 addr_key[ETH_ALEN] __aligned(2);
+	struct ieee80211_mesh_fast_tx_key key;
 
 	struct ieee80211_fast_tx fast_tx;
 	u8 hdr[sizeof(struct ieee80211s_hdr) + sizeof(rfc1042_header)];
@@ -333,7 +362,8 @@ void mesh_path_tx_root_frame(struct ieee80211_sub_if_data *sdata);
 
 bool mesh_action_is_path_sel(struct ieee80211_mgmt *mgmt);
 struct ieee80211_mesh_fast_tx *
-mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr);
+mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata,
+		 struct ieee80211_mesh_fast_tx_key *key);
 bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
 			      struct sk_buff *skb, u32 ctrl_flags);
 void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
net/mac80211/mesh_pathtbl.c +22 -9
index 91b55d6a68b97..a6b62169f0848 100644
@@ -37,8 +37,8 @@ static const struct rhashtable_params mesh_rht_params = {
 static const struct rhashtable_params fast_tx_rht_params = {
 	.nelem_hint = 10,
 	.automatic_shrinking = true,
-	.key_len = ETH_ALEN,
-	.key_offset = offsetof(struct ieee80211_mesh_fast_tx, addr_key),
+	.key_len = sizeof_field(struct ieee80211_mesh_fast_tx, key),
+	.key_offset = offsetof(struct ieee80211_mesh_fast_tx, key),
 	.head_offset = offsetof(struct ieee80211_mesh_fast_tx, rhash),
 	.hashfn = mesh_table_hash,
 };
@@ -431,20 +431,21 @@ static void mesh_fast_tx_entry_free(struct mesh_tx_cache *cache,
 }
 
 struct ieee80211_mesh_fast_tx *
-mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr)
+mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata,
+		 struct ieee80211_mesh_fast_tx_key *key)
 {
 	struct ieee80211_mesh_fast_tx *entry;
 	struct mesh_tx_cache *cache;
 
 	cache = &sdata->u.mesh.tx_cache;
-	entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params);
+	entry = rhashtable_lookup(&cache->rht, key, fast_tx_rht_params);
 	if (!entry)
 		return NULL;
 
 	if (!(entry->mpath->flags & MESH_PATH_ACTIVE) ||
 	    mpath_expired(entry->mpath)) {
 		spin_lock_bh(&cache->walk_lock);
-		entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params);
+		entry = rhashtable_lookup(&cache->rht, key, fast_tx_rht_params);
 		if (entry)
 		    mesh_fast_tx_entry_free(cache, entry);
 		spin_unlock_bh(&cache->walk_lock);
@@ -489,18 +490,24 @@ void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
 	if (!sta)
 		return;
 
+	build.key.type = MESH_FAST_TX_TYPE_LOCAL;
 	if ((meshhdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6) {
 		/* This is required to keep the mppath alive */
 		mppath = mpp_path_lookup(sdata, meshhdr->eaddr1);
 		if (!mppath)
 			return;
 		build.mppath = mppath;
+		if (!ether_addr_equal(meshhdr->eaddr2, sdata->vif.addr))
+			build.key.type = MESH_FAST_TX_TYPE_PROXIED;
 	} else if (ieee80211_has_a4(hdr->frame_control)) {
 		mppath = mpath;
 	} else {
 		return;
 	}
 
+	if (!ether_addr_equal(hdr->addr4, sdata->vif.addr))
+		build.key.type = MESH_FAST_TX_TYPE_FORWARDED;
+
 	/* rate limit, in case fast xmit can't be enabled */
 	if (mppath->fast_tx_check == jiffies)
 		return;
@@ -547,7 +554,7 @@ void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
 		}
 	}
 
-	memcpy(build.addr_key, mppath->dst, ETH_ALEN);
+	memcpy(build.key.addr, mppath->dst, ETH_ALEN);
 	build.timestamp = jiffies;
 	build.fast_tx.band = info->band;
 	build.fast_tx.da_offs = offsetof(struct ieee80211_hdr, addr3);
@@ -646,12 +653,18 @@ void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata,
 			     const u8 *addr)
 {
 	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
+	struct ieee80211_mesh_fast_tx_key key = {};
 	struct ieee80211_mesh_fast_tx *entry;
+	int i;
 
+	ether_addr_copy(key.addr, addr);
 	spin_lock_bh(&cache->walk_lock);
-	entry = rhashtable_lookup_fast(&cache->rht, addr, fast_tx_rht_params);
-	if (entry)
-		mesh_fast_tx_entry_free(cache, entry);
+	for (i = 0; i < NUM_MESH_FAST_TX_TYPE; i++) {
+		key.type = i;
+		entry = rhashtable_lookup_fast(&cache->rht, &key, fast_tx_rht_params);
+		if (entry)
+			mesh_fast_tx_entry_free(cache, entry);
+	}
 	spin_unlock_bh(&cache->walk_lock);
 }
 
net/mac80211/rx.c +10 -3
index 0bf72928ccfca..541b0f53c6c35 100644
@@ -2746,7 +2746,10 @@ ieee80211_rx_mesh_fast_forward(struct ieee80211_sub_if_data *sdata,
 			       struct sk_buff *skb, int hdrlen)
 {
 	struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh;
-	struct ieee80211_mesh_fast_tx *entry = NULL;
+	struct ieee80211_mesh_fast_tx_key key = {
+		.type = MESH_FAST_TX_TYPE_FORWARDED
+	};
+	struct ieee80211_mesh_fast_tx *entry;
 	struct ieee80211s_hdr *mesh_hdr;
 	struct tid_ampdu_tx *tid_tx;
 	struct sta_info *sta;
@@ -2755,9 +2758,13 @@ ieee80211_rx_mesh_fast_forward(struct ieee80211_sub_if_data *sdata,
 
 	mesh_hdr = (struct ieee80211s_hdr *)(skb->data + sizeof(eth));
 	if ((mesh_hdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6)
-		entry = mesh_fast_tx_get(sdata, mesh_hdr->eaddr1);
+		ether_addr_copy(key.addr, mesh_hdr->eaddr1);
 	else if (!(mesh_hdr->flags & MESH_FLAGS_AE))
-		entry = mesh_fast_tx_get(sdata, skb->data);
+		ether_addr_copy(key.addr, skb->data);
+	else
+		return false;
+
+	entry = mesh_fast_tx_get(sdata, &key);
 	if (!entry)
 		return false;
 
Signature · 1 line
2.43.0
author maintainer
#42 lore raw

PATCH 42/228 wifi: iwlwifi: mvm: remove old PASN station when adding a new one

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Avraham Stern <[email protected]>

[ Upstream commit dbfff5bf9292714f02ace002fea8ce6599ea1145 ]

If a PASN station is added, and an old PASN station already exists for the same mac address, remove the old station before adding the new one. Keeping the old station caueses old security context to be used in measurements.

---
 drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c +2 -0
index 233ae81884a0e..ae0eb585b61ee 100644
@@ -53,6 +53,8 @@ int iwl_mvm_ftm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	if (!pasn)
 		return -ENOBUFS;
 
+	iwl_mvm_ftm_remove_pasn_sta(mvm, addr);
+
 	pasn->cipher = iwl_mvm_cipher_to_location_cipher(cipher);
 
 	switch (pasn->cipher) {
Signature · 1 line
2.43.0
author maintainer
#43 lore raw

PATCH 43/228 wifi: iwlwifi: mvm: return uid from iwl_mvm_build_scan_cmd

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miri Korenblit <[email protected]>

[ Upstream commit bada85a3f584763deadd201147778c3e791d279c ]

This function is supposed to return a uid on success, and an errno in failure.
But it currently returns the return value of the specific cmd version handler, which in turn returns 0 on success and errno otherwise. This means that on success, iwl_mvm_build_scan_cmd will return 0 regardless if the actual uid.
Fix this by returning the uid if the handler succeeded.

---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/wireless/intel/iwlwifi/mvm/scan.c +2 -1
index 7b6f1cdca067e..97d44354dbbb5 100644
@@ -2815,7 +2815,8 @@ static int iwl_mvm_build_scan_cmd(struct iwl_mvm *mvm,
 		if (ver_handler->version != scan_ver)
 			continue;
 
-		return ver_handler->handler(mvm, vif, params, type, uid);
+		err = ver_handler->handler(mvm, vif, params, type, uid);
+		return err ? : uid;
 	}
 
 	err = iwl_mvm_scan_umac(mvm, vif, params, type, uid);
Signature · 1 line
2.43.0
author maintainer
#44 lore raw

PATCH 44/228 drm/gma500: Remove lid code

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Patrik Jakobsson <[email protected]>

[ Upstream commit 6aff4c26ed677b1f464f721fbd3e7767f24a684d ]

Due to a change in the order of initialization, the lid timer got started before proper setup was made. This resulted in a crash during boot.

The lid switch is handled by gma500 through a timer that periodically polls the opregion for changes. These types of ACPI events shouldn't be handled by the graphics driver so let's get rid of the lid code. This fixes the crash during boot.

---
 drivers/gpu/drm/gma500/Makefile     |  1 -
 drivers/gpu/drm/gma500/psb_device.c |  5 +-
 drivers/gpu/drm/gma500/psb_drv.h    |  9 ----
 drivers/gpu/drm/gma500/psb_lid.c    | 80 -----------------------------
 4 files changed, 1 insertion(+), 94 deletions(-)
 delete mode 100644 drivers/gpu/drm/gma500/psb_lid.c
drivers/gpu/drm/gma500/Makefile +0 -1
index 4f302cd5e1a6c..58fed80c7392a 100644
@@ -34,7 +34,6 @@ gma500_gfx-y += \
 	  psb_intel_lvds.o \
 	  psb_intel_modes.o \
 	  psb_intel_sdvo.o \
-	  psb_lid.o \
 	  psb_irq.o
 
 gma500_gfx-$(CONFIG_ACPI) +=  opregion.o
drivers/gpu/drm/gma500/psb_device.c +1 -4
index dcfcd7b89d4a1..6dece8f0e380f 100644
@@ -73,8 +73,7 @@ static int psb_backlight_setup(struct drm_device *dev)
 	}
 
 	psb_intel_lvds_set_brightness(dev, PSB_MAX_BRIGHTNESS);
-	/* This must occur after the backlight is properly initialised */
-	psb_lid_timer_init(dev_priv);
+
 	return 0;
 }
 
@@ -259,8 +258,6 @@ static int psb_chip_setup(struct drm_device *dev)
 
 static void psb_chip_teardown(struct drm_device *dev)
 {
-	struct drm_psb_private *dev_priv = to_drm_psb_private(dev);
-	psb_lid_timer_takedown(dev_priv);
 	gma_intel_teardown_gmbus(dev);
 }
 
drivers/gpu/drm/gma500/psb_drv.h +0 -9
index c5edfa4aa4ccd..83c17689c454f 100644
@@ -162,7 +162,6 @@
 #define PSB_NUM_VBLANKS 2
 
 #define PSB_WATCHDOG_DELAY (HZ * 2)
-#define PSB_LID_DELAY (HZ / 10)
 
 #define PSB_MAX_BRIGHTNESS		100
 
@@ -491,11 +490,7 @@ struct drm_psb_private {
 	/* Hotplug handling */
 	struct work_struct hotplug_work;
 
-	/* LID-Switch */
-	spinlock_t lid_lock;
-	struct timer_list lid_timer;
 	struct psb_intel_opregion opregion;
-	u32 lid_last_state;
 
 	/* Watchdog */
 	uint32_t apm_reg;
@@ -591,10 +586,6 @@ struct psb_ops {
 	int i2c_bus;		/* I2C bus identifier for Moorestown */
 };
 
-/* psb_lid.c */
-extern void psb_lid_timer_init(struct drm_psb_private *dev_priv);
-extern void psb_lid_timer_takedown(struct drm_psb_private *dev_priv);
-
 /* modesetting */
 extern void psb_modeset_init(struct drm_device *dev);
 extern void psb_modeset_cleanup(struct drm_device *dev);
drivers/gpu/drm/gma500/psb_lid.c deleted +0 -80

80 lines not on this page · raw message

Signature · 1 line
2.43.0
author maintainer
#45 lore raw

PATCH 45/228 wifi: mac80211_hwsim: init peer measurement result

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johannes Berg <[email protected]>

[ Upstream commit 2a4e01e5270b9fa9f6e6e0a4c24ac51a758636f9 ]

If we don't get all the values here, we might pass them to cfg80211 uninitialized. Fix that, even if the input might then not make much sense.

---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/virtual/mac80211_hwsim.c +1 -1
index a84340c2075ff..a9c3fb2ccdcdb 100644
@@ -3818,7 +3818,7 @@ static int hwsim_pmsr_report_nl(struct sk_buff *msg, struct genl_info *info)
 	}
 
 	nla_for_each_nested(peer, peers, rem) {
-		struct cfg80211_pmsr_result result;
+		struct cfg80211_pmsr_result result = {};
 
 		err = mac80211_hwsim_parse_pmsr_result(peer, &result, info);
 		if (err)
Signature · 1 line
2.43.0
author maintainer
#46 lore raw

PATCH 46/228 wifi: mac80211: remove link before AP

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johannes Berg <[email protected]>

[ Upstream commit cb55e08dba3526796e35d24a6d5db4ed6dcb8a4b ]

If the AP removal timer is long, we don't really want to remove the link immediately. However, we really should do it _before_ the AP removes it (which happens at or after count reaches 0), so subtract 1 from the countdown when scheduling the timer. This causes the link removal work to run just after the beacon with value 1 is received. If the counter is already zero, do it immediately.

This fixes an issue where we do the removal too late and receive a beacon from the AP that's no longer associated with the MLD, but thus removed EHT and ML elements, and then we disconnect instead from the whole MLD, since one of the associated APs changed mode from EHT to HE.

---
 net/mac80211/mlme.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
net/mac80211/mlme.c +5 -2
index 20d863370796d..fb9860d508707 100644
@@ -5873,8 +5873,11 @@ static void ieee80211_ml_reconfiguration(struct ieee80211_sub_if_data *sdata,
 			continue;
 		}
 
-		link_delay = link_conf->beacon_int *
-			link_removal_timeout[link_id];
+		if (link_removal_timeout[link_id] < 1)
+			link_delay = 0;
+		else
+			link_delay = link_conf->beacon_int *
+				(link_removal_timeout[link_id] - 1);
 
 		if (!delay)
 			delay = link_delay;
Signature · 1 line
2.43.0
author maintainer
#47 lore raw

PATCH 47/228 wifi: mac80211: fix unaligned le16 access

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johannes Berg <[email protected]>

[ Upstream commit c53d8a59351e4347452e263e2e5d7446ec93da83 ]

The AP removal timer field need not be aligned, so the code shouldn't access it directly, but use unaligned loads. Use get_unaligned_le16(), which even is shorter than the current code since it doesn't need a cast.

---
 net/mac80211/mlme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/mac80211/mlme.c +1 -1
index fb9860d508707..94028b541beba 100644
@@ -5848,7 +5848,7 @@ static void ieee80211_ml_reconfiguration(struct ieee80211_sub_if_data *sdata,
 		 */
 		if (control &
 		    IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT)
-			link_removal_timeout[link_id] = le16_to_cpu(*(__le16 *)pos);
+			link_removal_timeout[link_id] = get_unaligned_le16(pos);
 	}
 
 	removed_links &= sdata->vif.valid_links;
Signature · 1 line
2.43.0
author maintainer
#48 lore raw

PATCH 48/228 net: libwx: fix alloc msix vectors failed

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Duanqiang Wen <[email protected]>

[ Upstream commit 69197dfc64007b5292cc960581548f41ccd44828 ]

driver needs queue msix vectors and one misc irq vector, but only queue vectors need irq affinity.
when num_online_cpus is less than chip max msix vectors, driver will acquire (num_online_cpus + 1) vecotrs, and call pci_alloc_irq_vectors_affinity functions with affinity params without setting pre_vectors or post_vectors, it will cause return error code -ENOSPC.
Misc irq vector is vector 0, driver need to set affinity params .pre_vectors = 1.

Fixes 3f703186113f ("net: libwx: Add irq flow functions")
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -1
index 8706223a6e5aa..08d3e4069c5fa 100644
@@ -1598,7 +1598,7 @@ static void wx_set_num_queues(struct wx *wx)
  */
 static int wx_acquire_msix_vectors(struct wx *wx)
 {
-	struct irq_affinity affd = {0, };
+	struct irq_affinity affd = { .pre_vectors = 1 };
 	int nvecs, i;
 
 	/* We start by asking for one vector per queue pair */
Signature · 1 line
2.43.0
author maintainer
#49 lore raw

PATCH 49/228 vxlan: drop packets from invalid src-address

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Bauer <[email protected]>

[ Upstream commit f58f45c1e5b92975e91754f5407250085a6ae7cf ]

The VXLAN driver currently does not check if the inner layer2 source-address is valid.

In case source-address snooping/learning is enabled, a entry in the FDB for the invalid address is created with the layer3 address of the tunnel endpoint.

If the frame happens to have a non-unicast address set, all this non-unicast traffic is subsequently not flooded to the tunnel network but sent to the learnt host in the FDB. To make matters worse, this FDB entry does not expire.

Apply the same filtering for packets as it is done for bridges. This not only drops these invalid packets but avoids them from being learnt into the FDB.

Fixes d342894c5d2f ("vxlan: virtual extensible lan")
---
 drivers/net/vxlan/vxlan_core.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/net/vxlan/vxlan_core.c +4 -0
index 0bedb804824ba..9ec46048d361d 100644
@@ -1615,6 +1615,10 @@ static bool vxlan_set_mac(struct vxlan_dev *vxlan,
 	if (ether_addr_equal(eth_hdr(skb)->h_source, vxlan->dev->dev_addr))
 		return false;
 
+	/* Ignore packets from invalid src-address */
+	if (!is_valid_ether_addr(eth_hdr(skb)->h_source))
+		return false;
+
 	/* Get address from the outer IP header */
 	if (vxlan_get_sk_family(vs) == AF_INET) {
 		saddr.sin.sin_addr.s_addr = ip_hdr(skb)->saddr;
Signature · 1 line
2.43.0
author maintainer
#50 lore raw

PATCH 50/228 net: bcmasp: fix memory leak when bringing down interface

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Justin Chen <[email protected]>

[ Upstream commit 9f898fc2c31fbf0ac5ecd289f528a716464cb005 ]

When bringing down the TX rings we flush the rings but forget to reclaimed the flushed packets. This leads to a memory leak since we do not free the dma mapped buffers. This also leads to tx control block corruption when bringing down the interface for power management.

---
 .../net/ethernet/broadcom/asp2/bcmasp_intf.c  | 21 ++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)
drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c +14 -7
index 78901e2e73032..529172a87ae50 100644
@@ -435,10 +435,8 @@ static void umac_init(struct bcmasp_intf *intf)
 	umac_wl(intf, 0x800, UMC_RX_MAX_PKT_SZ);
 }
 
-static int bcmasp_tx_poll(struct napi_struct *napi, int budget)
+static int bcmasp_tx_reclaim(struct bcmasp_intf *intf)
 {
-	struct bcmasp_intf *intf =
-		container_of(napi, struct bcmasp_intf, tx_napi);
 	struct bcmasp_intf_stats64 *stats = &intf->stats64;
 	struct device *kdev = &intf->parent->pdev->dev;
 	unsigned long read, released = 0;
@@ -481,10 +479,16 @@ static int bcmasp_tx_poll(struct napi_struct *napi, int budget)
 							DESC_RING_COUNT);
 	}
 
-	/* Ensure all descriptors have been written to DRAM for the hardware
-	 * to see updated contents.
-	 */
-	wmb();
+	return released;
+}
+
+static int bcmasp_tx_poll(struct napi_struct *napi, int budget)
+{
+	struct bcmasp_intf *intf =
+		container_of(napi, struct bcmasp_intf, tx_napi);
+	int released = 0;
+
+	released = bcmasp_tx_reclaim(intf);
 
 	napi_complete(&intf->tx_napi);
 
@@ -796,6 +800,7 @@ static int bcmasp_init_tx(struct bcmasp_intf *intf)
 
 	intf->tx_spb_index = 0;
 	intf->tx_spb_clean_index = 0;
+	memset(intf->tx_cbs, 0, sizeof(struct bcmasp_tx_cb) * DESC_RING_COUNT);
 
 	netif_napi_add_tx(intf->ndev, &intf->tx_napi, bcmasp_tx_poll);
 
@@ -906,6 +911,8 @@ static void bcmasp_netif_deinit(struct net_device *dev)
 	} while (timeout-- > 0);
 	tx_spb_dma_wl(intf, 0x0, TX_SPB_DMA_FIFO_CTRL);
 
+	bcmasp_tx_reclaim(intf);
+
 	umac_enable_set(intf, UMC_CMD_TX_EN, 0);
 
 	phy_stop(dev->phydev);
Signature · 1 line
2.43.0
author maintainer
#51 lore raw

PATCH 51/228 mlxsw: core: Unregister EMAD trap using FORWARD action

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 976c44af48141cd8595601c0af2a19a43c5b228b ]

The device's manual (PRM - Programmer's Reference Manual) classifies the trap that is used to deliver EMAD responses as an "event trap". Among other things, it means that the only actions that can be associated with the trap are TRAP and FORWARD (NOP).

Currently, during driver de-initialization the driver unregisters the trap by setting its action to DISCARD, which violates the above guideline. Future firmware versions will prevent such misuses by returning an error. This does not prevent the driver from working, but an error will be printed to the kernel log during module removal / devlink reload:

mlxsw_spectrum 0000:03:00.0: Reg cmd access status failed (status=7(bad parameter)) mlxsw_spectrum 0000:03:00.0: Reg cmd access failed (reg_id=7003(hpkt),type=write)

Suppress the error message by aligning the driver to the manual and use a FORWARD (NOP) action when unregistering the trap.

---
 drivers/net/ethernet/mellanox/mlxsw/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/core.c +1 -1
index e4d7739bd7c88..4a79c0d7e7ad8 100644
@@ -849,7 +849,7 @@ static void mlxsw_emad_rx_listener_func(struct sk_buff *skb, u16 local_port,
 
 static const struct mlxsw_listener mlxsw_emad_rx_listener =
 	MLXSW_RXL(mlxsw_emad_rx_listener_func, ETHEMAD, TRAP_TO_CPU, false,
-		  EMAD, DISCARD);
+		  EMAD, FORWARD);
 
 static int mlxsw_emad_tlv_enable(struct mlxsw_core *mlxsw_core)
 {
Signature · 1 line
2.43.0
author maintainer
#52 lore raw

PATCH 52/228 mlxsw: core_env: Fix driver initialization with old firmware

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 7e2050a8366315aeaf0316b3d362e67cf58f3ea8 ]

The driver queries the Management Capabilities Mask (MCAM) register during initialization to understand if it can read up to 128 bytes from transceiver modules.

However, not all firmware versions support this register, leading to the driver failing to load.

Fix by treating an error in the register query as an indication that the feature is not supported.

---
 .../net/ethernet/mellanox/mlxsw/core_env.c    | 20 ++++++-------------
 1 file changed, 6 insertions(+), 14 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/core_env.c +6 -14
index 53b150b7ae4e7..6c06b05927608 100644
@@ -1357,24 +1357,20 @@ static struct mlxsw_linecards_event_ops mlxsw_env_event_ops = {
 	.got_inactive = mlxsw_env_got_inactive,
 };
 
-static int mlxsw_env_max_module_eeprom_len_query(struct mlxsw_env *mlxsw_env)
+static void mlxsw_env_max_module_eeprom_len_query(struct mlxsw_env *mlxsw_env)
 {
 	char mcam_pl[MLXSW_REG_MCAM_LEN];
-	bool mcia_128b_supported;
+	bool mcia_128b_supported = false;
 	int err;
 
 	mlxsw_reg_mcam_pack(mcam_pl,
 			    MLXSW_REG_MCAM_FEATURE_GROUP_ENHANCED_FEATURES);
 	err = mlxsw_reg_query(mlxsw_env->core, MLXSW_REG(mcam), mcam_pl);
-	if (err)
-		return err;
-
-	mlxsw_reg_mcam_unpack(mcam_pl, MLXSW_REG_MCAM_MCIA_128B,
-			      &mcia_128b_supported);
+	if (!err)
+		mlxsw_reg_mcam_unpack(mcam_pl, MLXSW_REG_MCAM_MCIA_128B,
+				      &mcia_128b_supported);
 
 	mlxsw_env->max_eeprom_len = mcia_128b_supported ? 128 : 48;
-
-	return 0;
 }
 
 int mlxsw_env_init(struct mlxsw_core *mlxsw_core,
@@ -1445,15 +1441,11 @@ int mlxsw_env_init(struct mlxsw_core *mlxsw_core,
 	if (err)
 		goto err_type_set;
 
-	err = mlxsw_env_max_module_eeprom_len_query(env);
-	if (err)
-		goto err_eeprom_len_query;
-
+	mlxsw_env_max_module_eeprom_len_query(env);
 	env->line_cards[0]->active = true;
 
 	return 0;
 
-err_eeprom_len_query:
 err_type_set:
 	mlxsw_env_module_event_disable(env, 0);
 err_mlxsw_env_module_event_enable:
Signature · 1 line
2.43.0
author maintainer
#53 lore raw

PATCH 53/228 mlxsw: pci: Fix driver initialization with old firmware

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 773501d01e6bc3f2557882a25679392d982d5f3e ]

The driver queries the Management Capabilities Mask (MCAM) register during initialization to understand if a new and deeper reset flow is supported.

However, not all firmware versions support this register, leading to the driver failing to load.

Fix by treating an error in the register query as an indication that the feature is not supported.

---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/pci.c +4 -6
index af99bf17eb36d..f42a1b1c93687 100644
@@ -1530,7 +1530,7 @@ mlxsw_pci_reset(struct mlxsw_pci *mlxsw_pci, const struct pci_device_id *id)
 {
 	struct pci_dev *pdev = mlxsw_pci->pdev;
 	char mcam_pl[MLXSW_REG_MCAM_LEN];
-	bool pci_reset_supported;
+	bool pci_reset_supported = false;
 	u32 sys_status;
 	int err;
 
@@ -1548,11 +1548,9 @@ mlxsw_pci_reset(struct mlxsw_pci *mlxsw_pci, const struct pci_device_id *id)
 	mlxsw_reg_mcam_pack(mcam_pl,
 			    MLXSW_REG_MCAM_FEATURE_GROUP_ENHANCED_FEATURES);
 	err = mlxsw_reg_query(mlxsw_pci->core, MLXSW_REG(mcam), mcam_pl);
-	if (err)
-		return err;
-
-	mlxsw_reg_mcam_unpack(mcam_pl, MLXSW_REG_MCAM_PCI_RESET,
-			      &pci_reset_supported);
+	if (!err)
+		mlxsw_reg_mcam_unpack(mcam_pl, MLXSW_REG_MCAM_PCI_RESET,
+				      &pci_reset_supported);
 
 	if (pci_reset_supported) {
 		pci_dbg(pdev, "Starting PCI reset flow\n");
Signature · 1 line
2.43.0
author maintainer
#54 lore raw

PATCH 54/228 ARM: dts: microchip: at91-sama7g5ek: Replace regulator-suspend-voltage with the valid property

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andrei Simion <[email protected]>

[ Upstream commit e027b71762e84ee9d4ba9ad5401b956b9e83ed2a ]

By checking the pmic node with microchip,mcp16502.yaml# 'regulator-suspend-voltage' does not match any of the regexes 'pinctrl-[0-9]+' from schema microchip,mcp16502.yaml# which inherits regulator.yaml#. So replace regulator-suspend-voltage with regulator-suspend-microvolt to avoid the inconsitency.

Fixes 85b1304b9daa ("ARM: dts: at91: sama7g5ek: set regulator voltages for standby state")

[claudiu.beznea: added a dot before starting the last sentence in commit
description]

---
 arch/arm/boot/dts/microchip/at91-sama7g5ek.dts | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
arch/arm/boot/dts/microchip/at91-sama7g5ek.dts +4 -4
index 217e9b96c61e5..20b2497657ae4 100644
@@ -293,7 +293,7 @@
 
 					regulator-state-standby {
 						regulator-on-in-suspend;
-						regulator-suspend-voltage = <1150000>;
+						regulator-suspend-microvolt = <1150000>;
 						regulator-mode = <4>;
 					};
 
@@ -314,7 +314,7 @@
 
 					regulator-state-standby {
 						regulator-on-in-suspend;
-						regulator-suspend-voltage = <1050000>;
+						regulator-suspend-microvolt = <1050000>;
 						regulator-mode = <4>;
 					};
 
@@ -331,7 +331,7 @@
 					regulator-always-on;
 
 					regulator-state-standby {
-						regulator-suspend-voltage = <1800000>;
+						regulator-suspend-microvolt = <1800000>;
 						regulator-on-in-suspend;
 					};
 
@@ -346,7 +346,7 @@
 					regulator-max-microvolt = <3700000>;
 
 					regulator-state-standby {
-						regulator-suspend-voltage = <1800000>;
+						regulator-suspend-microvolt = <1800000>;
 						regulator-on-in-suspend;
 					};
 
Signature · 1 line
2.43.0
author maintainer
#55 lore raw

PATCH 55/228 icmp: prevent possible NULL dereferences from icmp_build_probe()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit c58e88d49097bd12dfcfef4f075b43f5d5830941 ]

First problem is a double call to __in_dev_get_rcu(), because the second one could return NULL.

if (__in_dev_get_rcu(dev) && __in_dev_get_rcu(dev)->ifa_list)

Second problem is a read from dev->ip6_ptr with no NULL check:

if (!list_empty(&rcu_dereference(dev->ip6_ptr)->addr_list))

Use the correct RCU API to fix these.

v2: add missing include <net/addrconf.h>

Fixes d329ea5bd884 ("icmp: add response to RFC 8335 PROBE messages")
Cc Andreas Roeseler <[email protected]>
---
 net/ipv4/icmp.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)
net/ipv4/icmp.c +10 -2
index e63a3bf996176..437e782b9663b 100644
@@ -92,6 +92,7 @@
 #include <net/inet_common.h>
 #include <net/ip_fib.h>
 #include <net/l3mdev.h>
+#include <net/addrconf.h>
 
 /*
  *	Build xmit assembly blocks
@@ -1032,6 +1033,8 @@ bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr)
 	struct icmp_ext_hdr *ext_hdr, _ext_hdr;
 	struct icmp_ext_echo_iio *iio, _iio;
 	struct net *net = dev_net(skb->dev);
+	struct inet6_dev *in6_dev;
+	struct in_device *in_dev;
 	struct net_device *dev;
 	char buff[IFNAMSIZ];
 	u16 ident_len;
@@ -1115,10 +1118,15 @@ bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr)
 	/* Fill bits in reply message */
 	if (dev->flags & IFF_UP)
 		status |= ICMP_EXT_ECHOREPLY_ACTIVE;
-	if (__in_dev_get_rcu(dev) && __in_dev_get_rcu(dev)->ifa_list)
+
+	in_dev = __in_dev_get_rcu(dev);
+	if (in_dev && rcu_access_pointer(in_dev->ifa_list))
 		status |= ICMP_EXT_ECHOREPLY_IPV4;
-	if (!list_empty(&rcu_dereference(dev->ip6_ptr)->addr_list))
+
+	in6_dev = __in6_dev_get(dev);
+	if (in6_dev && !list_empty(&in6_dev->addr_list))
 		status |= ICMP_EXT_ECHOREPLY_IPV6;
+
 	dev_put(dev);
 	icmphdr->un.echo.sequence |= htons(status);
 	return true;
Signature · 1 line
2.43.0
author maintainer
#56 lore raw

PATCH 56/228 bridge/br_netlink.c: no need to return void function

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hangbin Liu <[email protected]>

[ Upstream commit 4fd1edcdf13c0d234543ecf502092be65c5177db ]

br_info_notify is a void function. There is no need to return.

Fixes b6d0425b816e ("bridge: cfm: Netlink Notifications.")
---
 net/bridge/br_netlink.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/bridge/br_netlink.c +1 -1
index 5ad4abfcb7ba3..d415833fce91f 100644
@@ -666,7 +666,7 @@ void br_ifinfo_notify(int event, const struct net_bridge *br,
 {
 	u32 filter = RTEXT_FILTER_BRVLAN_COMPRESSED;
 
-	return br_info_notify(event, br, port, filter);
+	br_info_notify(event, br, port, filter);
 }
 
 /*
Signature · 1 line
2.43.0
author maintainer
#57 lore raw

PATCH 57/228 bnxt_en: refactor reset close code

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vikas Gupta <[email protected]>

[ Upstream commit 7474b1c82be3780692d537d331f9aa7fc1e5a368 ]

Introduce bnxt_fw_fatal_close() API which can be used to stop data path and disable device when firmware is in fatal state.

Stable-dep-of: a1acdc226bae ("bnxt_en: Fix the PCI-AER routines")

---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +11 -6
index 5e6e32d708e24..260ffd71e2a89 100644
@@ -12776,6 +12776,16 @@ static void bnxt_rx_ring_reset(struct bnxt *bp)
 	bnxt_rtnl_unlock_sp(bp);
 }
 
+static void bnxt_fw_fatal_close(struct bnxt *bp)
+{
+	bnxt_tx_disable(bp);
+	bnxt_disable_napi(bp);
+	bnxt_disable_int_sync(bp);
+	bnxt_free_irq(bp);
+	bnxt_clear_int_mode(bp);
+	pci_disable_device(bp->pdev);
+}
+
 static void bnxt_fw_reset_close(struct bnxt *bp)
 {
 	bnxt_ulp_stop(bp);
@@ -12789,12 +12799,7 @@ static void bnxt_fw_reset_close(struct bnxt *bp)
 		pci_read_config_word(bp->pdev, PCI_SUBSYSTEM_ID, &val);
 		if (val == 0xffff)
 			bp->fw_reset_min_dsecs = 0;
-		bnxt_tx_disable(bp);
-		bnxt_disable_napi(bp);
-		bnxt_disable_int_sync(bp);
-		bnxt_free_irq(bp);
-		bnxt_clear_int_mode(bp);
-		pci_disable_device(bp->pdev);
+		bnxt_fw_fatal_close(bp);
 	}
 	__bnxt_close_nic(bp, true, false);
 	bnxt_vf_reps_free(bp);
Signature · 1 line
2.43.0
author maintainer
#58 lore raw

PATCH 58/228 bnxt_en: Fix the PCI-AER routines

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vikas Gupta <[email protected]>

[ Upstream commit a1acdc226baec331512f815d6ac9dd6f8435cc7f ]

We do not support two simultaneous recoveries so check for reset flag, BNXT_STATE_IN_FW_RESET, and do not proceed with AER further. When the pci channel state is pci_channel_io_frozen, the PCIe link can not be trusted so we disable the traffic immediately and stop BAR access by calling bnxt_fw_fatal_close(). BAR access after AER fatal error can cause an NMI.

Fixes f75d9a0aa967 ("bnxt_en: Re-write PCI BARs after PCI fatal error.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +16 -3
index 260ffd71e2a89..6bdd8c2607898 100644
@@ -15045,6 +15045,7 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
 {
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct bnxt *bp = netdev_priv(netdev);
+	bool abort = false;
 
 	netdev_info(netdev, "PCI I/O error detected\n");
 
@@ -15053,16 +15054,27 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
 
 	bnxt_ulp_stop(bp);
 
-	if (state == pci_channel_io_perm_failure) {
+	if (test_and_set_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		netdev_err(bp->dev, "Firmware reset already in progress\n");
+		abort = true;
+	}
+
+	if (abort || state == pci_channel_io_perm_failure) {
 		rtnl_unlock();
 		return PCI_ERS_RESULT_DISCONNECT;
 	}
 
-	if (state == pci_channel_io_frozen)
+	/* Link is not reliable anymore if state is pci_channel_io_frozen
+	 * so we disable bus master to prevent any potential bad DMAs before
+	 * freeing kernel memory.
+	 */
+	if (state == pci_channel_io_frozen) {
 		set_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, &bp->state);
+		bnxt_fw_fatal_close(bp);
+	}
 
 	if (netif_running(netdev))
-		bnxt_close(netdev);
+		__bnxt_close_nic(bp, true, true);
 
 	if (pci_is_enabled(pdev))
 		pci_disable_device(pdev);
@@ -15146,6 +15158,7 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
 	}
 
 reset_exit:
+	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
 	bnxt_clear_reservations(bp, true);
 	rtnl_unlock();
 
Signature · 1 line
2.43.0
author maintainer
#59 lore raw

PATCH 59/228 bnxt_en: Fix error recovery for 5760X (P7) chips

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Chan <[email protected]>

[ Upstream commit 41e54045b741daf61e03c82d442227af3d12111f ]

During error recovery, such as AER fatal error slot reset, we call bnxt_try_map_fw_health_reg() to try to get access to the health register to determine the firmware state. Fix bnxt_try_map_fw_health_reg() to recognize the P7 chip correctly and set up the health register.

This fixes this type of AER slot reset failure:

bnxt_en 0000:04:00.0: AER: PCIe Bus Error: severity=Uncorrectable (Fatal), type=Inaccessible, (Unregistered Agent ID) bnxt_en 0000:04:00.0 enp4s0f0np0: PCI I/O error detected
bnxt_en 0000:04:00.0 bnxt_re0: Handle device suspend call
bnxt_en 0000:04:00.1 enp4s0f1np1: PCI I/O error detected
bnxt_en 0000:04:00.1 bnxt_re1: Handle device suspend call
pcieport 0000:00:02.0: AER: Root Port link has been reset (0)
bnxt_en 0000:04:00.0 enp4s0f0np0: PCI Slot Reset
bnxt_en 0000:04:00.0: enabling device (0000 -> 0002)
bnxt_en 0000:04:00.0: Firmware not ready
bnxt_en 0000:04:00.1 enp4s0f1np1: PCI Slot Reset
bnxt_en 0000:04:00.1: enabling device (0000 -> 0002)
bnxt_en 0000:04:00.1: Firmware not ready
pcieport 0000:00:02.0: AER: device recovery failed

Fixes a432a45bdba4 ("bnxt_en: Define basic P7 macros")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +1 -1
index 6bdd8c2607898..5b4d810748f2a 100644
@@ -8906,7 +8906,7 @@ static void bnxt_try_map_fw_health_reg(struct bnxt *bp)
 					     BNXT_FW_HEALTH_WIN_BASE +
 					     BNXT_GRC_REG_CHIP_NUM);
 		}
-		if (!BNXT_CHIP_P5(bp))
+		if (!BNXT_CHIP_P5_PLUS(bp))
 			return;
 
 		status_loc = BNXT_GRC_REG_STATUS_P5 |
Signature · 1 line
2.43.0
author maintainer
#60 lore raw

PATCH 60/228 cxl/core: Fix potential payload size confusion in cxl_mem_get_poison()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dan Williams <[email protected]>

[ Upstream commit 4b759dd5765503bd466defac7d93aca14c23a15d ]

A recent change to cxl_mem_get_records_log() [1] highlighted a subtle nuance of looping calls to cxl_internal_send_cmd(), i.e. that cxl_internal_send_cmd() modifies the 'size_out' member of the @mbox_cmd argument. That mechanism is useful for communicating underflow, but it is unwanted when reusing @mbox_cmd for a subsequent submission. It turns out that cxl_xfer_log() avoids this scenario by always redefining @mbox_cmd each iteration.

Update cxl_mem_get_records_log() and cxl_mem_get_poison() to follow the same style as cxl_xfer_log(), i.e. re-define @mbox_cmd each iteration. The cxl_mem_get_records_log() change is just a style fixup, but the cxl_mem_get_poison() change is a potential fix, per Alison [2]:

Poison list retrieval can hit this case if the MORE flag is set and a follow on read of the list delivers more records than the previous read. ie. device gives one record, sets the _MORE flag, then gives 5.

Not an urgent fix since this behavior has not been seen in the wild, but worth tracking as a fix.

---
 drivers/cxl/core/mbox.c | 38 +++++++++++++++++---------------------
 1 file changed, 17 insertions(+), 21 deletions(-)
drivers/cxl/core/mbox.c +17 -21
index f0f54aeccc872..65185c9fa0013 100644
@@ -946,25 +946,22 @@ static void cxl_mem_get_records_log(struct cxl_memdev_state *mds,
 	struct cxl_memdev *cxlmd = mds->cxlds.cxlmd;
 	struct device *dev = mds->cxlds.dev;
 	struct cxl_get_event_payload *payload;
-	struct cxl_mbox_cmd mbox_cmd;
 	u8 log_type = type;
 	u16 nr_rec;
 
 	mutex_lock(&mds->event.log_lock);
 	payload = mds->event.buf;
 
-	mbox_cmd = (struct cxl_mbox_cmd) {
-		.opcode = CXL_MBOX_OP_GET_EVENT_RECORD,
-		.payload_in = &log_type,
-		.size_in = sizeof(log_type),
-		.payload_out = payload,
-		.min_out = struct_size(payload, records, 0),
-	};
-
 	do {
 		int rc, i;
-
-		mbox_cmd.size_out = mds->payload_size;
+		struct cxl_mbox_cmd mbox_cmd = (struct cxl_mbox_cmd) {
+			.opcode = CXL_MBOX_OP_GET_EVENT_RECORD,
+			.payload_in = &log_type,
+			.size_in = sizeof(log_type),
+			.payload_out = payload,
+			.size_out = mds->payload_size,
+			.min_out = struct_size(payload, records, 0),
+		};
 
 		rc = cxl_internal_send_cmd(mds, &mbox_cmd);
 		if (rc) {
@@ -1297,7 +1294,6 @@ int cxl_mem_get_poison(struct cxl_memdev *cxlmd, u64 offset, u64 len,
 	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
 	struct cxl_mbox_poison_out *po;
 	struct cxl_mbox_poison_in pi;
-	struct cxl_mbox_cmd mbox_cmd;
 	int nr_records = 0;
 	int rc;
 
@@ -1309,16 +1305,16 @@ int cxl_mem_get_poison(struct cxl_memdev *cxlmd, u64 offset, u64 len,
 	pi.offset = cpu_to_le64(offset);
 	pi.length = cpu_to_le64(len / CXL_POISON_LEN_MULT);
 
-	mbox_cmd = (struct cxl_mbox_cmd) {
-		.opcode = CXL_MBOX_OP_GET_POISON,
-		.size_in = sizeof(pi),
-		.payload_in = &pi,
-		.size_out = mds->payload_size,
-		.payload_out = po,
-		.min_out = struct_size(po, record, 0),
-	};
-
 	do {
+		struct cxl_mbox_cmd mbox_cmd = (struct cxl_mbox_cmd){
+			.opcode = CXL_MBOX_OP_GET_POISON,
+			.size_in = sizeof(pi),
+			.payload_in = &pi,
+			.size_out = mds->payload_size,
+			.payload_out = po,
+			.min_out = struct_size(po, record, 0),
+		};
+
 		rc = cxl_internal_send_cmd(mds, &mbox_cmd);
 		if (rc)
 			break;
Signature · 1 line
2.43.0
author maintainer
#61 lore raw

PATCH 61/228 net: dsa: mv88e6xx: fix supported_interfaces setup in mv88e6250_phylink_get_caps()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Matthias Schiffer <[email protected]>

[ Upstream commit a4e3899065ffa87d49dc20e8c17501edbc189692 ]

With the recent PHYLINK changes requiring supported_interfaces to be set, MV88E6250 family switches like the 88E6020 fail to probe - cmode is never initialized on these devices, so mv88e6250_phylink_get_caps() does not set any supported_interfaces flags.

Instead of a cmode, on 88E6250 we have a read-only port mode value that encodes similar information. There is no reason to bother mapping port mode to the cmodes of other switch models; instead we introduce a mv88e6250_setup_supported_interfaces() that is called directly from mv88e6250_phylink_get_caps().

Fixes de5c9bf40c45 ("net: phylink: require supported_interfaces to be filled")
---
 drivers/net/dsa/mv88e6xxx/chip.c | 56 +++++++++++++++++++++++++++++---
 drivers/net/dsa/mv88e6xxx/port.h | 23 ++++++++++---
 2 files changed, 71 insertions(+), 8 deletions(-)
drivers/net/dsa/mv88e6xxx/chip.c +52 -4
index 7de8a9c9e4576..32416d8802ca4 100644
@@ -566,13 +566,61 @@ static void mv88e6xxx_translate_cmode(u8 cmode, unsigned long *supported)
 		phy_interface_set_rgmii(supported);
 }
 
-static void mv88e6250_phylink_get_caps(struct mv88e6xxx_chip *chip, int port,
-				       struct phylink_config *config)
+static void
+mv88e6250_setup_supported_interfaces(struct mv88e6xxx_chip *chip, int port,
+				     struct phylink_config *config)
 {
 	unsigned long *supported = config->supported_interfaces;
+	int err;
+	u16 reg;
 
-	/* Translate the default cmode */
-	mv88e6xxx_translate_cmode(chip->ports[port].cmode, supported);
+	err = mv88e6xxx_port_read(chip, port, MV88E6XXX_PORT_STS, &reg);
+	if (err) {
+		dev_err(chip->dev, "p%d: failed to read port status\n", port);
+		return;
+	}
+
+	switch (reg & MV88E6250_PORT_STS_PORTMODE_MASK) {
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_HALF_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_100_HALF_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_FULL_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_100_FULL_PHY:
+		__set_bit(PHY_INTERFACE_MODE_REVMII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_HALF:
+	case MV88E6250_PORT_STS_PORTMODE_MII_FULL:
+		__set_bit(PHY_INTERFACE_MODE_MII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_200_RMII_FULL_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_HALF_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL_PHY:
+		__set_bit(PHY_INTERFACE_MODE_REVRMII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL:
+		__set_bit(PHY_INTERFACE_MODE_RMII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_100_RGMII:
+		__set_bit(PHY_INTERFACE_MODE_RGMII, supported);
+		break;
+
+	default:
+		dev_err(chip->dev,
+			"p%d: invalid port mode in status register: %04x\n",
+			port, reg);
+	}
+}
+
+static void mv88e6250_phylink_get_caps(struct mv88e6xxx_chip *chip, int port,
+				       struct phylink_config *config)
+{
+	if (!mv88e6xxx_phy_is_internal(chip, port))
+		mv88e6250_setup_supported_interfaces(chip, port, config);
 
 	config->mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100;
 }
drivers/net/dsa/mv88e6xxx/port.h +19 -4
index 86deeb347cbc1..ddadeb9bfdaee 100644
@@ -25,10 +25,25 @@
 #define MV88E6250_PORT_STS_PORTMODE_PHY_100_HALF	0x0900
 #define MV88E6250_PORT_STS_PORTMODE_PHY_10_FULL		0x0a00
 #define MV88E6250_PORT_STS_PORTMODE_PHY_100_FULL	0x0b00
-#define MV88E6250_PORT_STS_PORTMODE_MII_10_HALF		0x0c00
-#define MV88E6250_PORT_STS_PORTMODE_MII_100_HALF	0x0d00
-#define MV88E6250_PORT_STS_PORTMODE_MII_10_FULL		0x0e00
-#define MV88E6250_PORT_STS_PORTMODE_MII_100_FULL	0x0f00
+/* - Modes with PHY suffix use output instead of input clock
+ * - Modes without RMII or RGMII use MII
+ * - Modes without speed do not have a fixed speed specified in the manual
+ *   ("DC to x MHz" - variable clock support?)
+ */
+#define MV88E6250_PORT_STS_PORTMODE_MII_DISABLED		0x0000
+#define MV88E6250_PORT_STS_PORTMODE_MII_100_RGMII		0x0100
+#define MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL_PHY	0x0200
+#define MV88E6250_PORT_STS_PORTMODE_MII_200_RMII_FULL_PHY	0x0400
+#define MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL	0x0600
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL	0x0700
+#define MV88E6250_PORT_STS_PORTMODE_MII_HALF			0x0800
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_HALF_PHY	0x0900
+#define MV88E6250_PORT_STS_PORTMODE_MII_FULL			0x0a00
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL_PHY	0x0b00
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_HALF_PHY		0x0c00
+#define MV88E6250_PORT_STS_PORTMODE_MII_100_HALF_PHY		0x0d00
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_FULL_PHY		0x0e00
+#define MV88E6250_PORT_STS_PORTMODE_MII_100_FULL_PHY		0x0f00
 #define MV88E6XXX_PORT_STS_LINK			0x0800
 #define MV88E6XXX_PORT_STS_DUPLEX		0x0400
 #define MV88E6XXX_PORT_STS_SPEED_MASK		0x0300
Signature · 1 line
2.43.0
author maintainer
#62 lore raw

PATCH 62/228 NFC: trf7970a: disable all regulators on removal

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paul Geurts <[email protected]>

[ Upstream commit 6bea4f03c6a4e973ef369e15aac88f37981db49e ]

During module probe, regulator 'vin' and 'vdd-io' are used and enabled, but the vdd-io regulator overwrites the 'vin' regulator pointer. During remove, only the vdd-io is disabled, as the vin regulator pointer is not available anymore. When regulator_put() is called during resource cleanup a kernel warning is given, as the regulator is still enabled.

Store the two regulators in separate pointers and disable both the regulators on module remove.

---
 drivers/nfc/trf7970a.c | 42 +++++++++++++++++++++++-------------------
 1 file changed, 23 insertions(+), 19 deletions(-)
drivers/nfc/trf7970a.c +23 -19
index 7eb17f46a8153..9e1a34e23af26 100644
@@ -424,7 +424,8 @@ struct trf7970a {
 	enum trf7970a_state		state;
 	struct device			*dev;
 	struct spi_device		*spi;
-	struct regulator		*regulator;
+	struct regulator		*vin_regulator;
+	struct regulator		*vddio_regulator;
 	struct nfc_digital_dev		*ddev;
 	u32				quirks;
 	bool				is_initiator;
@@ -1883,7 +1884,7 @@ static int trf7970a_power_up(struct trf7970a *trf)
 	if (trf->state != TRF7970A_ST_PWR_OFF)
 		return 0;
 
-	ret = regulator_enable(trf->regulator);
+	ret = regulator_enable(trf->vin_regulator);
 	if (ret) {
 		dev_err(trf->dev, "%s - Can't enable VIN: %d\n", __func__, ret);
 		return ret;
@@ -1926,7 +1927,7 @@ static int trf7970a_power_down(struct trf7970a *trf)
 	if (trf->en2_gpiod && !(trf->quirks & TRF7970A_QUIRK_EN2_MUST_STAY_LOW))
 		gpiod_set_value_cansleep(trf->en2_gpiod, 0);
 
-	ret = regulator_disable(trf->regulator);
+	ret = regulator_disable(trf->vin_regulator);
 	if (ret)
 		dev_err(trf->dev, "%s - Can't disable VIN: %d\n", __func__,
 			ret);
@@ -2065,37 +2066,37 @@ static int trf7970a_probe(struct spi_device *spi)
 	mutex_init(&trf->lock);
 	INIT_DELAYED_WORK(&trf->timeout_work, trf7970a_timeout_work_handler);
 
-	trf->regulator = devm_regulator_get(&spi->dev, "vin");
-	if (IS_ERR(trf->regulator)) {
-		ret = PTR_ERR(trf->regulator);
+	trf->vin_regulator = devm_regulator_get(&spi->dev, "vin");
+	if (IS_ERR(trf->vin_regulator)) {
+		ret = PTR_ERR(trf->vin_regulator);
 		dev_err(trf->dev, "Can't get VIN regulator: %d\n", ret);
 		goto err_destroy_lock;
 	}
 
-	ret = regulator_enable(trf->regulator);
+	ret = regulator_enable(trf->vin_regulator);
 	if (ret) {
 		dev_err(trf->dev, "Can't enable VIN: %d\n", ret);
 		goto err_destroy_lock;
 	}
 
-	uvolts = regulator_get_voltage(trf->regulator);
+	uvolts = regulator_get_voltage(trf->vin_regulator);
 	if (uvolts > 4000000)
 		trf->chip_status_ctrl = TRF7970A_CHIP_STATUS_VRS5_3;
 
-	trf->regulator = devm_regulator_get(&spi->dev, "vdd-io");
-	if (IS_ERR(trf->regulator)) {
-		ret = PTR_ERR(trf->regulator);
+	trf->vddio_regulator = devm_regulator_get(&spi->dev, "vdd-io");
+	if (IS_ERR(trf->vddio_regulator)) {
+		ret = PTR_ERR(trf->vddio_regulator);
 		dev_err(trf->dev, "Can't get VDD_IO regulator: %d\n", ret);
-		goto err_destroy_lock;
+		goto err_disable_vin_regulator;
 	}
 
-	ret = regulator_enable(trf->regulator);
+	ret = regulator_enable(trf->vddio_regulator);
 	if (ret) {
 		dev_err(trf->dev, "Can't enable VDD_IO: %d\n", ret);
-		goto err_destroy_lock;
+		goto err_disable_vin_regulator;
 	}
 
-	if (regulator_get_voltage(trf->regulator) == 1800000) {
+	if (regulator_get_voltage(trf->vddio_regulator) == 1800000) {
 		trf->io_ctrl = TRF7970A_REG_IO_CTRL_IO_LOW;
 		dev_dbg(trf->dev, "trf7970a config vdd_io to 1.8V\n");
 	}
@@ -2108,7 +2109,7 @@ static int trf7970a_probe(struct spi_device *spi)
 	if (!trf->ddev) {
 		dev_err(trf->dev, "Can't allocate NFC digital device\n");
 		ret = -ENOMEM;
-		goto err_disable_regulator;
+		goto err_disable_vddio_regulator;
 	}
 
 	nfc_digital_set_parent_dev(trf->ddev, trf->dev);
@@ -2137,8 +2138,10 @@ static int trf7970a_probe(struct spi_device *spi)
 	trf7970a_shutdown(trf);
 err_free_ddev:
 	nfc_digital_free_device(trf->ddev);
-err_disable_regulator:
-	regulator_disable(trf->regulator);
+err_disable_vddio_regulator:
+	regulator_disable(trf->vddio_regulator);
+err_disable_vin_regulator:
+	regulator_disable(trf->vin_regulator);
 err_destroy_lock:
 	mutex_destroy(&trf->lock);
 	return ret;
@@ -2157,7 +2160,8 @@ static void trf7970a_remove(struct spi_device *spi)
 	nfc_digital_unregister_device(trf->ddev);
 	nfc_digital_free_device(trf->ddev);
 
-	regulator_disable(trf->regulator);
+	regulator_disable(trf->vddio_regulator);
+	regulator_disable(trf->vin_regulator);
 
 	mutex_destroy(&trf->lock);
 }
Signature · 1 line
2.43.0
author maintainer
#63 lore raw

PATCH 63/228 netfs: Fix writethrough-mode error handling

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Howells <[email protected]>

[ Upstream commit 619606a7b8d5e54b71578ecc988d3f8e1896bbc6 ]

Fix the error return in netfs_perform_write() acting in writethrough-mode to return any cached error in the case that netfs_end_writethrough() returns 0.

This can affect the use of O_SYNC/O_DSYNC/RWF_SYNC/RWF_DSYNC in 9p and afs.

Fixes 41d8e7673a77 ("netfs: Implement a write-through caching option")
cc Eric Van Hensbergen <[email protected]>
cc Latchesar Ionkov <[email protected]>
cc Dominique Martinet <[email protected]>
cc Christian Schoenebeck <[email protected]>
cc Marc Dionne <[email protected]>
---
 fs/netfs/buffered_write.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
fs/netfs/buffered_write.c +6 -4
index 9a0d32e4b422a..8f13ca8fbc74d 100644
@@ -164,7 +164,7 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter,
 	enum netfs_how_to_modify howto;
 	enum netfs_folio_trace trace;
 	unsigned int bdp_flags = (iocb->ki_flags & IOCB_SYNC) ? 0: BDP_ASYNC;
-	ssize_t written = 0, ret;
+	ssize_t written = 0, ret, ret2;
 	loff_t i_size, pos = iocb->ki_pos, from, to;
 	size_t max_chunk = PAGE_SIZE << MAX_PAGECACHE_ORDER;
 	bool maybe_trouble = false;
@@ -395,10 +395,12 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter,
 
 out:
 	if (unlikely(wreq)) {
-		ret = netfs_end_writethrough(wreq, iocb);
+		ret2 = netfs_end_writethrough(wreq, iocb);
 		wbc_detach_inode(&wbc);
-		if (ret == -EIOCBQUEUED)
-			return ret;
+		if (ret2 == -EIOCBQUEUED)
+			return ret2;
+		if (ret == 0)
+			ret = ret2;
 	}
 
 	iocb->ki_pos += written;
Signature · 1 line
2.43.0
author maintainer
#64 lore raw

PATCH 64/228 ax25: Fix netdev refcount issue

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Duoming Zhou <[email protected]>

[ Upstream commit 467324bcfe1a31ec65d0cf4aa59421d6b7a7d52b ]

The dev_tracker is added to ax25_cb in ax25_bind(). When the ax25 device is detaching, the dev_tracker of ax25_cb should be deallocated in ax25_kill_by_device() instead of the dev_tracker of ax25_dev. The log reported by ref_tracker is shown below:

[   80.884935] ref_tracker: reference already released.
[   80.885150] ref_tracker: allocated in:
[   80.885349]  ax25_dev_device_up+0x105/0x540
[   80.885730]  ax25_device_event+0xa4/0x420
[   80.885730]  notifier_call_chain+0xc9/0x1e0
[   80.885730]  __dev_notify_flags+0x138/0x280
[   80.885730]  dev_change_flags+0xd7/0x180
[   80.885730]  dev_ifsioc+0x6a9/0xa30
[   80.885730]  dev_ioctl+0x4d8/0xd90
[   80.885730]  sock_do_ioctl+0x1c2/0x2d0
[   80.885730]  sock_ioctl+0x38b/0x4f0
[   80.885730]  __se_sys_ioctl+0xad/0xf0
[   80.885730]  do_syscall_64+0xc4/0x1b0
[   80.885730]  entry_SYSCALL_64_after_hwframe+0x67/0x6f
[   80.885730] ref_tracker: freed in:
[   80.885730]  ax25_device_event+0x272/0x420
[   80.885730]  notifier_call_chain+0xc9/0x1e0
[   80.885730]  dev_close_many+0x272/0x370
[   80.885730]  unregister_netdevice_many_notify+0x3b5/0x1180
[   80.885730]  unregister_netdev+0xcf/0x120
[   80.885730]  sixpack_close+0x11f/0x1b0
[   80.885730]  tty_ldisc_kill+0xcb/0x190
[   80.885730]  tty_ldisc_hangup+0x338/0x3d0
[   80.885730]  __tty_hangup+0x504/0x740
[   80.885730]  tty_release+0x46e/0xd80
[   80.885730]  __fput+0x37f/0x770
[   80.885730]  __x64_sys_close+0x7b/0xb0
[   80.885730]  do_syscall_64+0xc4/0x1b0
[   80.885730]  entry_SYSCALL_64_after_hwframe+0x67/0x6f
[   80.893739] ------------[ cut here ]------------
[   80.894030] WARNING: CPU: 2 PID: 140 at lib/ref_tracker.c:255 ref_tracker_free+0x47b/0x6b0
[   80.894297] Modules linked in:
[   80.894929] CPU: 2 PID: 140 Comm: ax25_conn_rel_6 Not tainted 6.9.0-rc4-g8cd26fd90c1a #11
[   80.895190] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qem4
[   80.895514] RIP: 0010:ref_tracker_free+0x47b/0x6b0
[   80.895808] Code: 83 c5 18 4c 89 eb 48 c1 eb 03 8a 04 13 84 c0 0f 85 df 01 00 00 41 83 7d 00 00 75 4b 4c 89 ff 9
[   80.896171] RSP: 0018:ffff888009edf8c0 EFLAGS: 00000286
[   80.896339] RAX: 1ffff1100141ac00 RBX: 1ffff1100149463b RCX: dffffc0000000000
[   80.896502] RDX: 0000000000000001 RSI: 0000000000000246 RDI: ffff88800a0d6518
[   80.896925] RBP: ffff888009edf9b0 R08: ffff88806d3288d3 R09: 1ffff1100da6511a
[   80.897212] R10: dffffc0000000000 R11: ffffed100da6511b R12: ffff88800a4a31d4
[   80.897859] R13: ffff88800a4a31d8 R14: dffffc0000000000 R15: ffff88800a0d6518
[   80.898279] FS:  00007fd88b7fe700(0000) GS:ffff88806d300000(0000) knlGS:0000000000000000
[   80.899436] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   80.900181] CR2: 00007fd88c001d48 CR3: 000000000993e000 CR4: 00000000000006f0
...
[   80.935774] ref_tracker: sp%d@000000000bb9df3d has 1/1 users at
[   80.935774]      ax25_bind+0x424/0x4e0
[   80.935774]      __sys_bind+0x1d9/0x270
[   80.935774]      __x64_sys_bind+0x75/0x80
[   80.935774]      do_syscall_64+0xc4/0x1b0
[   80.935774]      entry_SYSCALL_64_after_hwframe+0x67/0x6f

Change ax25_dev->dev_tracker to the dev_tracker of ax25_cb in order to mitigate the bug.

Fixes feef318c855a ("ax25: fix UAF bugs of net_device caused by rebinding operation")
---
 net/ax25/af_ax25.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ax25/af_ax25.c +1 -1
index 558e158c98d01..9169efb2f43aa 100644
@@ -103,7 +103,7 @@ static void ax25_kill_by_device(struct net_device *dev)
 			s->ax25_dev = NULL;
 			if (sk->sk_socket) {
 				netdev_put(ax25_dev->dev,
-					   &ax25_dev->dev_tracker);
+					   &s->dev_tracker);
 				ax25_dev_put(ax25_dev);
 			}
 			ax25_cb_del(s);
Signature · 1 line
2.43.0
author maintainer
#65 lore raw

PATCH 65/228 soc: mediatek: mtk-svs: Append "-thermal" to thermal zone names

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: AngeloGioacchino Del Regno <[email protected]>

[ Upstream commit 7ca803b489455b9242c81b4befe546ea3a692e5c ]

The thermal framework registers thermal zones as specified in DT and including the "-thermal" suffix: append that to the driver specified tzone_name to actually match the thermal zone name as registered by the thermal API.

Fixes 2bfbf82956e2 ("soc: mediatek: mtk-svs: Constify runtime-immutable members of svs_bank")
---
 drivers/soc/mediatek/mtk-svs.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/soc/mediatek/mtk-svs.c +5 -2
index c832f5c670bcf..9a91298c12539 100644
@@ -1768,6 +1768,7 @@ static int svs_bank_resource_setup(struct svs_platform *svsp)
 	const struct svs_bank_pdata *bdata;
 	struct svs_bank *svsb;
 	struct dev_pm_opp *opp;
+	char tz_name_buf[20];
 	unsigned long freq;
 	int count, ret;
 	u32 idx, i;
@@ -1819,10 +1820,12 @@ static int svs_bank_resource_setup(struct svs_platform *svsp)
 		}
 
 		if (!IS_ERR_OR_NULL(bdata->tzone_name)) {
-			svsb->tzd = thermal_zone_get_zone_by_name(bdata->tzone_name);
+			snprintf(tz_name_buf, ARRAY_SIZE(tz_name_buf),
+				 "%s-thermal", bdata->tzone_name);
+			svsb->tzd = thermal_zone_get_zone_by_name(tz_name_buf);
 			if (IS_ERR(svsb->tzd)) {
 				dev_err(svsb->dev, "cannot get \"%s\" thermal zone\n",
-					bdata->tzone_name);
+					tz_name_buf);
 				return PTR_ERR(svsb->tzd);
 			}
 		}
Signature · 1 line
2.43.0
author maintainer
#66 lore raw

PATCH 66/228 tools: ynl: dont ignore errors in NLMSG_DONE messages

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jakub Kicinski <[email protected]>

[ Upstream commit a44f2eb106a46f2275a79de54ce0ea63e4f3d8c8 ]

NLMSG_DONE contains an error code, it has to be extracted. Prior to this change all dumps will end in success, and in case of failure the result is silently truncated.

Fixes e4b48ed460d3 ("tools: ynl: add a completely generic client")
---
 tools/net/ynl/lib/ynl.py | 1 +
 1 file changed, 1 insertion(+)
tools/net/ynl/lib/ynl.py +1 -0
index 1e10512b2117c..571c2e218993f 100644
@@ -208,6 +208,7 @@ class NlMsg:
             self.done = 1
             extack_off = 20
         elif self.nl_type == Netlink.NLMSG_DONE:
+            self.error = struct.unpack("i", self.raw[0:4])[0]
             self.done = 1
             extack_off = 4
 
Signature · 1 line
2.43.0
author maintainer
#67 lore raw

PATCH 67/228 net: make SK_MEMORY_PCPU_RESERV tunable

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Adam Li <[email protected]>

[ Upstream commit 12a686c2e761f1f1f6e6e2117a9ab9c6de2ac8a7 ]

This patch adds /proc/sys/net/core/mem_pcpu_rsv sysctl file, to make SK_MEMORY_PCPU_RESERV tunable.

Commit 3cd3399dd7a8 ("net: implement per-cpu reserves for memory_allocated") introduced per-cpu forward alloc cache:

"Implement a per-cpu cache of +1/-1 MB, to reduce number of changes to sk->sk_prot->memory_allocated, which would otherwise be cause of false sharing."

sk_prot->memory_allocated points to global atomic variable:
atomic_long_t tcp_memory_allocated ____cacheline_aligned_in_smp;

If increasing the per-cpu cache size from 1MB to e.g. 16MB, changes to sk->sk_prot->memory_allocated can be further reduced. Performance may be improved on system with many cores.

Stable-dep-of: 3584718cf2ec ("net: fix sk_memory_allocated_{add|sub} vs softirqs")

---
 Documentation/admin-guide/sysctl/net.rst | 5 +++++
 include/net/sock.h                       | 5 +++--
 net/core/sock.c                          | 1 +
 net/core/sysctl_net_core.c               | 9 +++++++++
 4 files changed, 18 insertions(+), 2 deletions(-)
Documentation/admin-guide/sysctl/net.rst +5 -0
index 3960916519557..7250c0542828b 100644
@@ -206,6 +206,11 @@ Will increase power usage.
 
 Default: 0 (off)
 
+mem_pcpu_rsv
+------------
+
+Per-cpu reserved forward alloc cache size in page units. Default 1MB per CPU.
+
 rmem_default
 ------------
 
include/net/sock.h +3 -2
index 4afb4376c97e8..6855c50a70f1b 100644
@@ -1429,6 +1429,7 @@ sk_memory_allocated(const struct sock *sk)
 
 /* 1 MB per cpu, in page units */
 #define SK_MEMORY_PCPU_RESERVE (1 << (20 - PAGE_SHIFT))
+extern int sysctl_mem_pcpu_rsv;
 
 static inline void
 sk_memory_allocated_add(struct sock *sk, int amt)
@@ -1437,7 +1438,7 @@ sk_memory_allocated_add(struct sock *sk, int amt)
 
 	preempt_disable();
 	local_reserve = __this_cpu_add_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve >= SK_MEMORY_PCPU_RESERVE) {
+	if (local_reserve >= READ_ONCE(sysctl_mem_pcpu_rsv)) {
 		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
 		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
 	}
@@ -1451,7 +1452,7 @@ sk_memory_allocated_sub(struct sock *sk, int amt)
 
 	preempt_disable();
 	local_reserve = __this_cpu_sub_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve <= -SK_MEMORY_PCPU_RESERVE) {
+	if (local_reserve <= -READ_ONCE(sysctl_mem_pcpu_rsv)) {
 		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
 		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
 	}
net/core/sock.c +1 -0
index 5e78798456fd8..9cf404e8038a4 100644
@@ -283,6 +283,7 @@ __u32 sysctl_rmem_max __read_mostly = SK_RMEM_MAX;
 EXPORT_SYMBOL(sysctl_rmem_max);
 __u32 sysctl_wmem_default __read_mostly = SK_WMEM_MAX;
 __u32 sysctl_rmem_default __read_mostly = SK_RMEM_MAX;
+int sysctl_mem_pcpu_rsv __read_mostly = SK_MEMORY_PCPU_RESERVE;
 
 int sysctl_tstamp_allow_data __read_mostly = 1;
 
net/core/sysctl_net_core.c +9 -0
index 0f0cb1465e089..986f15e5d6c41 100644
@@ -30,6 +30,7 @@ static int int_3600 = 3600;
 static int min_sndbuf = SOCK_MIN_SNDBUF;
 static int min_rcvbuf = SOCK_MIN_RCVBUF;
 static int max_skb_frags = MAX_SKB_FRAGS;
+static int min_mem_pcpu_rsv = SK_MEMORY_PCPU_RESERVE;
 
 static int net_msg_warn;	/* Unused, but still a sysctl */
 
@@ -407,6 +408,14 @@ static struct ctl_table net_core_table[] = {
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= &min_rcvbuf,
 	},
+	{
+		.procname	= "mem_pcpu_rsv",
+		.data		= &sysctl_mem_pcpu_rsv,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_mem_pcpu_rsv,
+	},
 	{
 		.procname	= "dev_weight",
 		.data		= &weight_p,
Signature · 1 line
2.43.0
author maintainer
#68 lore raw

PATCH 68/228 net: fix sk_memory_allocated_{add|sub} vs softirqs

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 3584718cf2ec7e79b6814f2596dcf398c5fb2eca ]

Jonathan Heathcote reported a regression caused by blamed commit on aarch64 architecture.

x86 happens to have irq-safe __this_cpu_add_return() and __this_cpu_sub(), but this is not generic.

I think my confusion came from "struct sock" argument, because these helpers are called with a locked socket. But the memory accounting is per-proto (and per-cpu after the blamed commit). We might cleanup these helpers later to directly accept a "struct proto *proto" argument.

Switch to this_cpu_add_return() and this_cpu_xchg() operations, and get rid of preempt_disable()/preempt_enable() pairs.

Fast path becomes a bit faster as a result :)

Many thanks to Jonathan Heathcote for his awesome report and investigations.

---
 include/net/sock.h | 38 ++++++++++++++++++++------------------
 1 file changed, 20 insertions(+), 18 deletions(-)
include/net/sock.h +20 -18
index 6855c50a70f1b..54a7967613348 100644
@@ -1431,32 +1431,34 @@ sk_memory_allocated(const struct sock *sk)
 #define SK_MEMORY_PCPU_RESERVE (1 << (20 - PAGE_SHIFT))
 extern int sysctl_mem_pcpu_rsv;
 
+static inline void proto_memory_pcpu_drain(struct proto *proto)
+{
+	int val = this_cpu_xchg(*proto->per_cpu_fw_alloc, 0);
+
+	if (val)
+		atomic_long_add(val, proto->memory_allocated);
+}
+
 static inline void
-sk_memory_allocated_add(struct sock *sk, int amt)
+sk_memory_allocated_add(const struct sock *sk, int val)
 {
-	int local_reserve;
+	struct proto *proto = sk->sk_prot;
 
-	preempt_disable();
-	local_reserve = __this_cpu_add_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve >= READ_ONCE(sysctl_mem_pcpu_rsv)) {
-		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
-		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
-	}
-	preempt_enable();
+	val = this_cpu_add_return(*proto->per_cpu_fw_alloc, val);
+
+	if (unlikely(val >= READ_ONCE(sysctl_mem_pcpu_rsv)))
+		proto_memory_pcpu_drain(proto);
 }
 
 static inline void
-sk_memory_allocated_sub(struct sock *sk, int amt)
+sk_memory_allocated_sub(const struct sock *sk, int val)
 {
-	int local_reserve;
+	struct proto *proto = sk->sk_prot;
 
-	preempt_disable();
-	local_reserve = __this_cpu_sub_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve <= -READ_ONCE(sysctl_mem_pcpu_rsv)) {
-		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
-		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
-	}
-	preempt_enable();
+	val = this_cpu_sub_return(*proto->per_cpu_fw_alloc, val);
+
+	if (unlikely(val <= -READ_ONCE(sysctl_mem_pcpu_rsv)))
+		proto_memory_pcpu_drain(proto);
 }
 
 #define SK_ALLOC_PERCPU_COUNTER_BATCH 16
Signature · 1 line
2.43.0
author maintainer
#69 lore raw

PATCH 69/228 ipv4: check for NULL idev in ip_route_use_hint()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 58a4c9b1e5a3e53c9148e80b90e1e43897ce77d1 ]

syzbot was able to trigger a NULL deref in fib_validate_source() in an old tree [1].

It appears the bug exists in latest trees.

All calls to __in_dev_get_rcu() must be checked for a NULL result.

[1]

general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
CPU: 2 PID: 3257 Comm: syz-executor.3 Not tainted 5.10.0-syzkaller #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
 RIP: 0010:fib_validate_source+0xbf/0x15a0 net/ipv4/fib_frontend.c:425
Code: 18 f2 f2 f2 f2 42 c7 44 20 23 f3 f3 f3 f3 48 89 44 24 78 42 c6 44 20 27 f3 e8 5d 88 48 fc 4c 89 e8 48 c1 e8 03 48 89 44 24 18 <42> 80 3c 20 00 74 08 4c 89 ef e8 d2 15 98 fc 48 89 5c 24 10 41 bf
RSP: 0018:ffffc900015fee40 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff88800f7a4000 RCX: ffff88800f4f90c0
RDX: 0000000000000000 RSI: 0000000004001eac RDI: ffff8880160c64c0
RBP: ffffc900015ff060 R08: 0000000000000000 R09: ffff88800f7a4000
R10: 0000000000000002 R11: ffff88800f4f90c0 R12: dffffc0000000000
R13: 0000000000000000 R14: 0000000000000000 R15: ffff88800f7a4000
FS:  00007f938acfe6c0(0000) GS:ffff888058c00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f938acddd58 CR3: 000000001248e000 CR4: 0000000000352ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
  ip_route_use_hint+0x410/0x9b0 net/ipv4/route.c:2231
  ip_rcv_finish_core+0x2c4/0x1a30 net/ipv4/ip_input.c:327
  ip_list_rcv_finish net/ipv4/ip_input.c:612 [inline]
  ip_sublist_rcv+0x3ed/0xe50 net/ipv4/ip_input.c:638
  ip_list_rcv+0x422/0x470 net/ipv4/ip_input.c:673
  __netif_receive_skb_list_ptype net/core/dev.c:5572 [inline]
  __netif_receive_skb_list_core+0x6b1/0x890 net/core/dev.c:5620
  __netif_receive_skb_list net/core/dev.c:5672 [inline]
  netif_receive_skb_list_internal+0x9f9/0xdc0 net/core/dev.c:5764
  netif_receive_skb_list+0x55/0x3e0 net/core/dev.c:5816
  xdp_recv_frames net/bpf/test_run.c:257 [inline]
  xdp_test_run_batch net/bpf/test_run.c:335 [inline]
  bpf_test_run_xdp_live+0x1818/0x1d00 net/bpf/test_run.c:363
  bpf_prog_test_run_xdp+0x81f/0x1170 net/bpf/test_run.c:1376
  bpf_prog_test_run+0x349/0x3c0 kernel/bpf/syscall.c:3736
  __sys_bpf+0x45c/0x710 kernel/bpf/syscall.c:5115
  __do_sys_bpf kernel/bpf/syscall.c:5201 [inline]
  __se_sys_bpf kernel/bpf/syscall.c:5199 [inline]
  __x64_sys_bpf+0x7c/0x90 kernel/bpf/syscall.c:5199
---
 net/ipv4/route.c | 3 +++
 1 file changed, 3 insertions(+)
net/ipv4/route.c +3 -0
index 15c37c8113fc8..f67d3d6fe9345 100644
@@ -2166,6 +2166,9 @@ int ip_route_use_hint(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	int err = -EINVAL;
 	u32 tag = 0;
 
+	if (!in_dev)
+		return -EINVAL;
+
 	if (ipv4_is_multicast(saddr) || ipv4_is_lbcast(saddr))
 		goto martian_source;
 
Signature · 1 line
2.43.0
author maintainer
#70 lore raw

PATCH 70/228 net: usb: ax88179_178a: stop lying about skb->truesize

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 4ce62d5b2f7aecd4900e7d6115588ad7f9acccca ]

Some usb drivers try to set small skb->truesize and break core networking stacks.

In this patch, I removed one of the skb->truesize overide.

I also replaced one skb_clone() by an allocation of a fresh and small skb, to get minimally sized skbs, like we did in commit 1e2c61172342 ("net: cdc_ncm: reduce skb truesize in rx path")

---
 drivers/net/usb/ax88179_178a.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)
drivers/net/usb/ax88179_178a.c +3 -8
index 3078511f76083..21b6c4d94a632 100644
@@ -1456,21 +1456,16 @@ static int ax88179_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 			/* Skip IP alignment pseudo header */
 			skb_pull(skb, 2);
 
-			skb->truesize = SKB_TRUESIZE(pkt_len_plus_padd);
 			ax88179_rx_checksum(skb, pkt_hdr);
 			return 1;
 		}
 
-		ax_skb = skb_clone(skb, GFP_ATOMIC);
+		ax_skb = netdev_alloc_skb_ip_align(dev->net, pkt_len);
 		if (!ax_skb)
 			return 0;
-		skb_trim(ax_skb, pkt_len);
+		skb_put(ax_skb, pkt_len);
+		memcpy(ax_skb->data, skb->data + 2, pkt_len);
 
-		/* Skip IP alignment pseudo header */
-		skb_pull(ax_skb, 2);
-
-		skb->truesize = pkt_len_plus_padd +
-				SKB_DATA_ALIGN(sizeof(struct sk_buff));
 		ax88179_rx_checksum(ax_skb, pkt_hdr);
 		usbnet_skb_return(dev, ax_skb);
 
Signature · 1 line
2.43.0
author maintainer
#71 lore raw

PATCH 71/228 tcp: Fix Use-After-Free in tcp_ao_connect_init

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit 80e679b352c3ce5158f3f778cfb77eb767e586fb ]

Since call_rcu, which is called in the hlist_for_each_entry_rcu traversal of tcp_ao_connect_init, is not part of the RCU read critical section, it is possible that the RCU grace period will pass during the traversal and the key will be free.

To prevent this, it should be changed to hlist_for_each_entry_safe.

---
 net/ipv4/tcp_ao.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/ipv4/tcp_ao.c +2 -1
index 87db432c6bb4a..254d6e3f93fa7 100644
@@ -1068,6 +1068,7 @@ void tcp_ao_connect_init(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct tcp_ao_info *ao_info;
+	struct hlist_node *next;
 	union tcp_ao_addr *addr;
 	struct tcp_ao_key *key;
 	int family, l3index;
@@ -1090,7 +1091,7 @@ void tcp_ao_connect_init(struct sock *sk)
 	l3index = l3mdev_master_ifindex_by_index(sock_net(sk),
 						 sk->sk_bound_dev_if);
 
-	hlist_for_each_entry_rcu(key, &ao_info->head, node) {
+	hlist_for_each_entry_safe(key, next, &ao_info->head, node) {
 		if (!tcp_ao_key_cmp(key, l3index, addr, key->prefixlen, family, -1, -1))
 			continue;
 
Signature · 1 line
2.43.0
author maintainer
#72 lore raw

PATCH 72/228 net: gtp: Fix Use-After-Free in gtp_dellink

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit f2a904107ee2b647bb7794a1a82b67740d7c8a64 ]

Since call_rcu, which is called in the hlist_for_each_entry_rcu traversal of gtp_dellink, is not part of the RCU read critical section, it is possible that the RCU grace period will pass during the traversal and the key will be free.

To prevent this, it should be changed to hlist_for_each_entry_safe.

Fixes 94dc550a5062 ("gtp: fix an use-after-free in ipv4_pdp_find()")
---
 drivers/net/gtp.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/gtp.c +2 -1
index 2b5357d94ff56..63f932256c9f5 100644
@@ -1111,11 +1111,12 @@ static int gtp_newlink(struct net *src_net, struct net_device *dev,
 static void gtp_dellink(struct net_device *dev, struct list_head *head)
 {
 	struct gtp_dev *gtp = netdev_priv(dev);
+	struct hlist_node *next;
 	struct pdp_ctx *pctx;
 	int i;
 
 	for (i = 0; i < gtp->hash_size; i++)
-		hlist_for_each_entry_rcu(pctx, &gtp->tid_hash[i], hlist_tid)
+		hlist_for_each_entry_safe(pctx, next, &gtp->tid_hash[i], hlist_tid)
 			pdp_context_delete(pctx);
 
 	list_del_rcu(&gtp->list);
Signature · 1 line
2.43.0
author maintainer
#73 lore raw

PATCH 73/228 net: phy: mediatek-ge-soc: follow netdev LED trigger semantics

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Daniel Golle <[email protected]>

[ Upstream commit 5b5f724b05c550e10693a53a81cadca901aefd16 ]

Only blink if the link is up on a LED which is programmed to also indicate link-status.

Otherwise, if both LEDs are in use to indicate different speeds, the resulting blinking being inverted on LEDs which aren't switched on at a specific speed is quite counter-intuitive.

Also make sure that state left behind by reset or the bootloader is recognized correctly including the half-duplex and full-duplex bits as well as the (unsupported by Linux netdev trigger semantics) link-down bit.

Fixes c66937b0f8db ("net: phy: mediatek-ge-soc: support PHY LEDs")
---
 drivers/net/phy/mediatek-ge-soc.c | 43 +++++++++++++++++++------------
 1 file changed, 26 insertions(+), 17 deletions(-)
drivers/net/phy/mediatek-ge-soc.c +26 -17
index 0f3a1538a8b8e..f4f9412d0cd7e 100644
@@ -216,6 +216,9 @@
 #define   MTK_PHY_LED_ON_LINK1000		BIT(0)
 #define   MTK_PHY_LED_ON_LINK100		BIT(1)
 #define   MTK_PHY_LED_ON_LINK10			BIT(2)
+#define   MTK_PHY_LED_ON_LINK			(MTK_PHY_LED_ON_LINK10 |\
+						 MTK_PHY_LED_ON_LINK100 |\
+						 MTK_PHY_LED_ON_LINK1000)
 #define   MTK_PHY_LED_ON_LINKDOWN		BIT(3)
 #define   MTK_PHY_LED_ON_FDX			BIT(4) /* Full duplex */
 #define   MTK_PHY_LED_ON_HDX			BIT(5) /* Half duplex */
@@ -231,6 +234,12 @@
 #define   MTK_PHY_LED_BLINK_100RX		BIT(3)
 #define   MTK_PHY_LED_BLINK_10TX		BIT(4)
 #define   MTK_PHY_LED_BLINK_10RX		BIT(5)
+#define   MTK_PHY_LED_BLINK_RX			(MTK_PHY_LED_BLINK_10RX |\
+						 MTK_PHY_LED_BLINK_100RX |\
+						 MTK_PHY_LED_BLINK_1000RX)
+#define   MTK_PHY_LED_BLINK_TX			(MTK_PHY_LED_BLINK_10TX |\
+						 MTK_PHY_LED_BLINK_100TX |\
+						 MTK_PHY_LED_BLINK_1000TX)
 #define   MTK_PHY_LED_BLINK_COLLISION		BIT(6)
 #define   MTK_PHY_LED_BLINK_RX_CRC_ERR		BIT(7)
 #define   MTK_PHY_LED_BLINK_RX_IDLE_ERR		BIT(8)
@@ -1247,11 +1256,9 @@ static int mt798x_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
 	if (blink < 0)
 		return -EIO;
 
-	if ((on & (MTK_PHY_LED_ON_LINK1000 | MTK_PHY_LED_ON_LINK100 |
-		   MTK_PHY_LED_ON_LINK10)) ||
-	    (blink & (MTK_PHY_LED_BLINK_1000RX | MTK_PHY_LED_BLINK_100RX |
-		      MTK_PHY_LED_BLINK_10RX | MTK_PHY_LED_BLINK_1000TX |
-		      MTK_PHY_LED_BLINK_100TX | MTK_PHY_LED_BLINK_10TX)))
+	if ((on & (MTK_PHY_LED_ON_LINK | MTK_PHY_LED_ON_FDX | MTK_PHY_LED_ON_HDX |
+		   MTK_PHY_LED_ON_LINKDOWN)) ||
+	    (blink & (MTK_PHY_LED_BLINK_RX | MTK_PHY_LED_BLINK_TX)))
 		set_bit(bit_netdev, &priv->led_state);
 	else
 		clear_bit(bit_netdev, &priv->led_state);
@@ -1269,7 +1276,7 @@ static int mt798x_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
 	if (!rules)
 		return 0;
 
-	if (on & (MTK_PHY_LED_ON_LINK1000 | MTK_PHY_LED_ON_LINK100 | MTK_PHY_LED_ON_LINK10))
+	if (on & MTK_PHY_LED_ON_LINK)
 		*rules |= BIT(TRIGGER_NETDEV_LINK);
 
 	if (on & MTK_PHY_LED_ON_LINK10)
@@ -1287,10 +1294,10 @@ static int mt798x_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
 	if (on & MTK_PHY_LED_ON_HDX)
 		*rules |= BIT(TRIGGER_NETDEV_HALF_DUPLEX);
 
-	if (blink & (MTK_PHY_LED_BLINK_1000RX | MTK_PHY_LED_BLINK_100RX | MTK_PHY_LED_BLINK_10RX))
+	if (blink & MTK_PHY_LED_BLINK_RX)
 		*rules |= BIT(TRIGGER_NETDEV_RX);
 
-	if (blink & (MTK_PHY_LED_BLINK_1000TX | MTK_PHY_LED_BLINK_100TX | MTK_PHY_LED_BLINK_10TX))
+	if (blink & MTK_PHY_LED_BLINK_TX)
 		*rules |= BIT(TRIGGER_NETDEV_TX);
 
 	return 0;
@@ -1323,15 +1330,19 @@ static int mt798x_phy_led_hw_control_set(struct phy_device *phydev, u8 index,
 		on |= MTK_PHY_LED_ON_LINK1000;
 
 	if (rules & BIT(TRIGGER_NETDEV_RX)) {
-		blink |= MTK_PHY_LED_BLINK_10RX  |
-			 MTK_PHY_LED_BLINK_100RX |
-			 MTK_PHY_LED_BLINK_1000RX;
+		blink |= (on & MTK_PHY_LED_ON_LINK) ?
+			  (((on & MTK_PHY_LED_ON_LINK10) ? MTK_PHY_LED_BLINK_10RX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK100) ? MTK_PHY_LED_BLINK_100RX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK1000) ? MTK_PHY_LED_BLINK_1000RX : 0)) :
+			  MTK_PHY_LED_BLINK_RX;
 	}
 
 	if (rules & BIT(TRIGGER_NETDEV_TX)) {
-		blink |= MTK_PHY_LED_BLINK_10TX  |
-			 MTK_PHY_LED_BLINK_100TX |
-			 MTK_PHY_LED_BLINK_1000TX;
+		blink |= (on & MTK_PHY_LED_ON_LINK) ?
+			  (((on & MTK_PHY_LED_ON_LINK10) ? MTK_PHY_LED_BLINK_10TX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK100) ? MTK_PHY_LED_BLINK_100TX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK1000) ? MTK_PHY_LED_BLINK_1000TX : 0)) :
+			  MTK_PHY_LED_BLINK_TX;
 	}
 
 	if (blink || on)
@@ -1344,9 +1355,7 @@ static int mt798x_phy_led_hw_control_set(struct phy_device *phydev, u8 index,
 				MTK_PHY_LED0_ON_CTRL,
 			     MTK_PHY_LED_ON_FDX     |
 			     MTK_PHY_LED_ON_HDX     |
-			     MTK_PHY_LED_ON_LINK10  |
-			     MTK_PHY_LED_ON_LINK100 |
-			     MTK_PHY_LED_ON_LINK1000,
+			     MTK_PHY_LED_ON_LINK,
 			     on);
 
 	if (ret)
Signature · 1 line
2.43.0
author maintainer
#74 lore raw

PATCH 74/228 gpio: tegra186: Fix tegra186_gpio_is_accessible() check

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Prathamesh Shete <[email protected]>

[ Upstream commit d806f474a9a7993648a2c70642ee129316d8deff ]

The controller has several register bits describing access control information for a given GPIO pin. When SCR_SEC_[R|W]EN is unset, it means we have full read/write access to all the registers for given GPIO pin. When SCR_SEC[R|W]EN is set, it means we need to further check the accompanying SCR_SEC_G1[R|W] bit to determine read/write access to all the registers for given GPIO pin.

This check was previously declaring that a GPIO pin was accessible only if either of the following conditions were met:

- SCR_SEC_REN + SCR_SEC_WEN both set

or

- SCR_SEC_REN + SCR_SEC_WEN both set and
SCR_SEC_G1R + SCR_SEC_G1W both set

Update the check to properly handle cases where only one of SCR_SEC_REN or SCR_SEC_WEN is set.

Fixes b2b56a163230 ("gpio: tegra186: Check GPIO pin permission before access.")
---
 drivers/gpio/gpio-tegra186.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)
drivers/gpio/gpio-tegra186.c +11 -9
index d87dd06db40d0..9130c691a2dd3 100644
@@ -36,12 +36,6 @@
 #define  TEGRA186_GPIO_SCR_SEC_REN		BIT(27)
 #define  TEGRA186_GPIO_SCR_SEC_G1W		BIT(9)
 #define  TEGRA186_GPIO_SCR_SEC_G1R		BIT(1)
-#define  TEGRA186_GPIO_FULL_ACCESS		(TEGRA186_GPIO_SCR_SEC_WEN | \
-						 TEGRA186_GPIO_SCR_SEC_REN | \
-						 TEGRA186_GPIO_SCR_SEC_G1R | \
-						 TEGRA186_GPIO_SCR_SEC_G1W)
-#define  TEGRA186_GPIO_SCR_SEC_ENABLE		(TEGRA186_GPIO_SCR_SEC_WEN | \
-						 TEGRA186_GPIO_SCR_SEC_REN)
 
 /* control registers */
 #define TEGRA186_GPIO_ENABLE_CONFIG 0x00
@@ -177,10 +171,18 @@ static inline bool tegra186_gpio_is_accessible(struct tegra_gpio *gpio, unsigned
 
 	value = __raw_readl(secure + TEGRA186_GPIO_SCR);
 
-	if ((value & TEGRA186_GPIO_SCR_SEC_ENABLE) == 0)
-		return true;
+	/*
+	 * When SCR_SEC_[R|W]EN is unset, then we have full read/write access to all the
+	 * registers for given GPIO pin.
+	 * When SCR_SEC[R|W]EN is set, then there is need to further check the accompanying
+	 * SCR_SEC_G1[R|W] bit to determine read/write access to all the registers for given
+	 * GPIO pin.
+	 */
 
-	if ((value & TEGRA186_GPIO_FULL_ACCESS) == TEGRA186_GPIO_FULL_ACCESS)
+	if (((value & TEGRA186_GPIO_SCR_SEC_REN) == 0 ||
+	     ((value & TEGRA186_GPIO_SCR_SEC_REN) && (value & TEGRA186_GPIO_SCR_SEC_G1R))) &&
+	     ((value & TEGRA186_GPIO_SCR_SEC_WEN) == 0 ||
+	     ((value & TEGRA186_GPIO_SCR_SEC_WEN) && (value & TEGRA186_GPIO_SCR_SEC_G1W))))
 		return true;
 
 	return false;
Signature · 1 line
2.43.0
author maintainer
#75 lore raw

PATCH 75/228 drm/xe: Remove sysfs only once on action add failure

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Himal Prasad Ghimiray <[email protected]>

[ Upstream commit d6dab9017b7cf155e73ba5c7f498de1beb5f8e24 ]

The drmm_add_action_or_reset function automatically invokes the action (sysfs removal) in the event of a failure; therefore, there's no necessity to call it within the return check.

Modify the return type of xe_gt_ccs_mode_sysfs_init to int, allowing the caller to pass errors up the call chain. Should sysfs creation or drmm_add_action_or_reset fail, error propagation will prompt a driver load abort.

-v2
Edit commit message (Nikula/Lucas)
use err_force_wake label instead of new. (Lucas)
Avoid unnecessary warn/error messages. (Lucas)

Fixes f3bc5bb4d53d ("drm/xe: Allow userspace to configure CCS mode")
Cc Lucas De Marchi <[email protected]>
Cc Jani Nikula <[email protected]>
Cc Rodrigo Vivi <[email protected]>
Cc Niranjana Vishwanathapura <[email protected]>
(cherry picked from commit a99641e38704202ae2a97202b3d249208c9cda7f)
---
 drivers/gpu/drm/xe/xe_gt.c          |  4 +++-
 drivers/gpu/drm/xe/xe_gt_ccs_mode.c | 19 +++++++------------
 drivers/gpu/drm/xe/xe_gt_ccs_mode.h |  2 +-
 3 files changed, 11 insertions(+), 14 deletions(-)
drivers/gpu/drm/xe/xe_gt.c +3 -1
index 35474ddbaf97e..d23ee1397cf07 100644
@@ -384,7 +384,9 @@ static int gt_fw_domain_init(struct xe_gt *gt)
 			 err);
 
 	/* Initialize CCS mode sysfs after early initialization of HW engines */
-	xe_gt_ccs_mode_sysfs_init(gt);
+	err = xe_gt_ccs_mode_sysfs_init(gt);
+	if (err)
+		goto err_force_wake;
 
 	err = xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
 	XE_WARN_ON(err);
drivers/gpu/drm/xe/xe_gt_ccs_mode.c +7 -12
index 529fc286cd06c..396aeb5b99242 100644
@@ -167,25 +167,20 @@ static void xe_gt_ccs_mode_sysfs_fini(struct drm_device *drm, void *arg)
  * and it is expected that there are no open drm clients while doing so.
  * The number of available compute slices is exposed to user through a per-gt
  * 'num_cslices' sysfs interface.
+ *
+ * Returns: Returns error value for failure and 0 for success.
  */
-void xe_gt_ccs_mode_sysfs_init(struct xe_gt *gt)
+int xe_gt_ccs_mode_sysfs_init(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 	int err;
 
 	if (!xe_gt_ccs_mode_enabled(gt))
-		return;
+		return 0;
 
 	err = sysfs_create_files(gt->sysfs, gt_ccs_mode_attrs);
-	if (err) {
-		drm_warn(&xe->drm, "Sysfs creation for ccs_mode failed err: %d\n", err);
-		return;
-	}
+	if (err)
+		return err;
 
-	err = drmm_add_action_or_reset(&xe->drm, xe_gt_ccs_mode_sysfs_fini, gt);
-	if (err) {
-		sysfs_remove_files(gt->sysfs, gt_ccs_mode_attrs);
-		drm_warn(&xe->drm, "%s: drmm_add_action_or_reset failed, err: %d\n",
-			 __func__, err);
-	}
+	return drmm_add_action_or_reset(&xe->drm, xe_gt_ccs_mode_sysfs_fini, gt);
 }
drivers/gpu/drm/xe/xe_gt_ccs_mode.h +1 -1
index f39975aaaab0d..f8779852cf0d2 100644
@@ -12,7 +12,7 @@
 #include "xe_platform_types.h"
 
 void xe_gt_apply_ccs_mode(struct xe_gt *gt);
-void xe_gt_ccs_mode_sysfs_init(struct xe_gt *gt);
+int xe_gt_ccs_mode_sysfs_init(struct xe_gt *gt);
 
 static inline bool xe_gt_ccs_mode_enabled(const struct xe_gt *gt)
 {
Signature · 1 line
2.43.0
author maintainer
#76 lore raw

PATCH 76/228 drm/xe: call free_gsc_pkt only once on action add failure

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Himal Prasad Ghimiray <[email protected]>

[ Upstream commit f38c4d224aa37fce1e3fe05db4377ef888f0737f ]

The drmm_add_action_or_reset function automatically invokes the action (free_gsc_pkt) in the event of a failure; therefore, there's no necessity to call it within the return check.

-v2
Fix commit message. (Lucas)

Fixes d8b1571312b7 ("drm/xe/huc: HuC authentication via GSC")
Cc Rodrigo Vivi <[email protected]>
Cc Daniele Ceraolo Spurio <[email protected]>
(cherry picked from commit 22bf0bc04d273ca002a47de55693797b13076602)
---
 drivers/gpu/drm/xe/xe_huc.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)
drivers/gpu/drm/xe/xe_huc.c +1 -8
index eca109791c6ae..01b2d0bd26a30 100644
@@ -53,7 +53,6 @@ static int huc_alloc_gsc_pkt(struct xe_huc *huc)
 	struct xe_gt *gt = huc_to_gt(huc);
 	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_bo *bo;
-	int err;
 
 	/* we use a single object for both input and output */
 	bo = xe_bo_create_pin_map(xe, gt_to_tile(gt), NULL,
@@ -66,13 +65,7 @@ static int huc_alloc_gsc_pkt(struct xe_huc *huc)
 
 	huc->gsc_pkt = bo;
 
-	err = drmm_add_action_or_reset(&xe->drm, free_gsc_pkt, huc);
-	if (err) {
-		free_gsc_pkt(&xe->drm, huc);
-		return err;
-	}
-
-	return 0;
+	return drmm_add_action_or_reset(&xe->drm, free_gsc_pkt, huc);
 }
 
 int xe_huc_init(struct xe_huc *huc)
Signature · 1 line
2.43.0
author maintainer
#77 lore raw

PATCH 77/228 Bluetooth: hci_event: Use HCI error defines instead of magic values

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jonas Dreßler <[email protected]>

[ Upstream commit 79c0868ad65a8fc7cdfaa5f2b77a4b70d0b0ea16 ]

We have error defines already, so let's use them.

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci.h | 2 ++
 net/bluetooth/hci_event.c   | 8 ++++----
 2 files changed, 6 insertions(+), 4 deletions(-)
include/net/bluetooth/hci.h +2 -0
index 35c5f75a3a5ee..04ec9c6e69e46 100644
@@ -668,6 +668,7 @@ enum {
 #define HCI_ERROR_PIN_OR_KEY_MISSING	0x06
 #define HCI_ERROR_MEMORY_EXCEEDED	0x07
 #define HCI_ERROR_CONNECTION_TIMEOUT	0x08
+#define HCI_ERROR_COMMAND_DISALLOWED	0x0c
 #define HCI_ERROR_REJ_LIMITED_RESOURCES	0x0d
 #define HCI_ERROR_REJ_BAD_ADDR		0x0f
 #define HCI_ERROR_INVALID_PARAMETERS	0x12
@@ -676,6 +677,7 @@ enum {
 #define HCI_ERROR_REMOTE_POWER_OFF	0x15
 #define HCI_ERROR_LOCAL_HOST_TERM	0x16
 #define HCI_ERROR_PAIRING_NOT_ALLOWED	0x18
+#define HCI_ERROR_UNSUPPORTED_REMOTE_FEATURE	0x1e
 #define HCI_ERROR_INVALID_LL_PARAMS	0x1e
 #define HCI_ERROR_UNSPECIFIED		0x1f
 #define HCI_ERROR_ADVERTISING_TIMEOUT	0x3c
net/bluetooth/hci_event.c +4 -4
index 3992e18f0babb..a59b5b4711a46 100644
@@ -93,11 +93,11 @@ static u8 hci_cc_inquiry_cancel(struct hci_dev *hdev, void *data,
 	/* It is possible that we receive Inquiry Complete event right
 	 * before we receive Inquiry Cancel Command Complete event, in
 	 * which case the latter event should have status of Command
-	 * Disallowed (0x0c). This should not be treated as error, since
+	 * Disallowed. This should not be treated as error, since
 	 * we actually achieve what Inquiry Cancel wants to achieve,
 	 * which is to end the last Inquiry session.
 	 */
-	if (rp->status == 0x0c && !test_bit(HCI_INQUIRY, &hdev->flags)) {
+	if (rp->status == HCI_ERROR_COMMAND_DISALLOWED && !test_bit(HCI_INQUIRY, &hdev->flags)) {
 		bt_dev_warn(hdev, "Ignoring error of Inquiry Cancel command");
 		rp->status = 0x00;
 	}
@@ -2340,7 +2340,7 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status)
 
 	if (status) {
 		if (conn && conn->state == BT_CONNECT) {
-			if (status != 0x0c || conn->attempt > 2) {
+			if (status != HCI_ERROR_COMMAND_DISALLOWED || conn->attempt > 2) {
 				conn->state = BT_CLOSED;
 				hci_connect_cfm(conn, status);
 				hci_conn_del(conn);
@@ -6710,7 +6710,7 @@ static void hci_le_remote_feat_complete_evt(struct hci_dev *hdev, void *data,
 			 * transition into connected state and mark it as
 			 * successful.
 			 */
-			if (!conn->out && ev->status == 0x1a &&
+			if (!conn->out && ev->status == HCI_ERROR_UNSUPPORTED_REMOTE_FEATURE &&
 			    (hdev->le_features[0] & HCI_LE_PERIPHERAL_FEATURES))
 				status = 0x00;
 			else
Signature · 1 line
2.43.0
author maintainer
#78 lore raw

PATCH 78/228 Bluetooth: hci_conn: Only do ACL connections sequentially

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jonas Dreßler <[email protected]>

[ Upstream commit 45340097ce6ea7e875674a5a7d24c95ecbc93ef9 ]

Pretty much all bluetooth chipsets only support paging a single device at a time, and if they don't reject a secondary "Create Connection" request while another is still ongoing, they'll most likely serialize those requests in the firware.

With commit 4c67bc74f016 ("[Bluetooth] Support concurrent connect requests") we started adding some serialization of our own in case the adapter returns "Command Disallowed" HCI error.

This commit was using the BT_CONNECT2 state for the serialization, this state is also used for a few more things (most notably to indicate we're waiting for an inquiry to cancel) and therefore a bit unreliable. Also not all BT firwares would respond with "Command Disallowed" on too many connection requests, some will also respond with "Hardware Failure" (BCM4378), and others will error out later and send a "Connect Complete" event with error "Rejected Limited Resources" (Marvell 88W8897).

We can clean things up a bit and also make the serialization more reliable by using our hci_sync machinery to always do "Create Connection" requests in a sequential manner.

This is very similar to what we're already doing for establishing LE connections, and it works well there.

Note that this causes a test failure in mgmt-tester (test "Pair Device
- Power off 1") because the hci_abort_conn_sync() changes the error we return on timeout of the "Create Connection". We'll fix this on the mgmt-tester side by adjusting the expected error for the test.

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci.h      |  1 +
 include/net/bluetooth/hci_sync.h |  3 ++
 net/bluetooth/hci_conn.c         | 69 ++++---------------------------
 net/bluetooth/hci_sync.c         | 70 ++++++++++++++++++++++++++++++++
 4 files changed, 83 insertions(+), 60 deletions(-)
include/net/bluetooth/hci.h +1 -0
index 04ec9c6e69e46..bfddb648fcc3e 100644
@@ -453,6 +453,7 @@ enum {
 #define HCI_NCMD_TIMEOUT	msecs_to_jiffies(4000)	/* 4 seconds */
 #define HCI_ACL_TX_TIMEOUT	msecs_to_jiffies(45000)	/* 45 seconds */
 #define HCI_AUTO_OFF_TIMEOUT	msecs_to_jiffies(2000)	/* 2 seconds */
+#define HCI_ACL_CONN_TIMEOUT	msecs_to_jiffies(20000)	/* 20 seconds */
 #define HCI_LE_CONN_TIMEOUT	msecs_to_jiffies(20000)	/* 20 seconds */
 #define HCI_LE_AUTOCONN_TIMEOUT	msecs_to_jiffies(4000)	/* 4 seconds */
 
include/net/bluetooth/hci_sync.h +3 -0
index e2582c2425449..824660f8f30da 100644
@@ -138,3 +138,6 @@ int hci_le_terminate_big_sync(struct hci_dev *hdev, u8 handle, u8 reason);
 int hci_le_big_terminate_sync(struct hci_dev *hdev, u8 handle);
 
 int hci_le_pa_terminate_sync(struct hci_dev *hdev, u16 handle);
+
+int hci_acl_create_connection_sync(struct hci_dev *hdev,
+				   struct hci_conn *conn);
net/bluetooth/hci_conn.c +9 -60

90 lines not on this page · raw message

net/bluetooth/hci_sync.c +70 -0

73 lines not on this page · raw message

Signature · 1 line
2.43.0
author maintainer
#79 lore raw

PATCH 79/228 Bluetooth: Remove pending ACL connection attempts

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jonas Dreßler <[email protected]>

[ Upstream commit 4aa42119d971603dc9e4d8cf4f53d5fcf082ea7d ]

With the last commit we moved to using the hci_sync queue for "Create Connection" requests, removing the need for retrying the paging after finished/failed "Create Connection" requests and after the end of inquiries.

hci_conn_check_pending() was used to trigger this retry, we can remove it now.

Note that we can also remove the special handling for COMMAND_DISALLOWED errors in the completion handler of "Create Connection", because "Create Connection" requests are now always serialized.

This is somewhat reverting commit 4c67bc74f016 ("[Bluetooth] Support concurrent connect requests").

With this, the BT_CONNECT2 state of ACL hci_conn objects should now be back to meaning only one thing: That we received a "Connection Request" from another device (see hci_conn_request_evt), but the response to that is going to be deferred.

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci_core.h |  1 -
 net/bluetooth/hci_conn.c         | 16 ----------------
 net/bluetooth/hci_event.c        | 21 ++++-----------------
 3 files changed, 4 insertions(+), 34 deletions(-)
include/net/bluetooth/hci_core.h +0 -1
index d1228857b1d0f..ddbf287ee64fb 100644
@@ -1481,7 +1481,6 @@ struct hci_conn *hci_conn_add_unset(struct hci_dev *hdev, int type,
 				    bdaddr_t *dst, u8 role);
 void hci_conn_del(struct hci_conn *conn);
 void hci_conn_hash_flush(struct hci_dev *hdev);
-void hci_conn_check_pending(struct hci_dev *hdev);
 
 struct hci_chan *hci_chan_create(struct hci_conn *conn);
 void hci_chan_del(struct hci_chan *chan);
net/bluetooth/hci_conn.c +0 -16
index d21e39ae67658..13481f08e47ee 100644
@@ -2592,22 +2592,6 @@ void hci_conn_hash_flush(struct hci_dev *hdev)
 	}
 }
 
-/* Check pending connect attempts */
-void hci_conn_check_pending(struct hci_dev *hdev)
-{
-	struct hci_conn *conn;
-
-	BT_DBG("hdev %s", hdev->name);
-
-	hci_dev_lock(hdev);
-
-	conn = hci_conn_hash_lookup_state(hdev, ACL_LINK, BT_CONNECT2);
-	if (conn)
-		hci_acl_create_connection_sync(hdev, conn);
-
-	hci_dev_unlock(hdev);
-}
-
 static u32 get_link_mode(struct hci_conn *conn)
 {
 	u32 link_mode = 0;
net/bluetooth/hci_event.c +4 -17
index a59b5b4711a46..f5d77c8089d81 100644
@@ -118,8 +118,6 @@ static u8 hci_cc_inquiry_cancel(struct hci_dev *hdev, void *data,
 		hci_discovery_set_state(hdev, DISCOVERY_STOPPED);
 	hci_dev_unlock(hdev);
 
-	hci_conn_check_pending(hdev);
-
 	return rp->status;
 }
 
@@ -150,8 +148,6 @@ static u8 hci_cc_exit_periodic_inq(struct hci_dev *hdev, void *data,
 
 	hci_dev_clear_flag(hdev, HCI_PERIODIC_INQ);
 
-	hci_conn_check_pending(hdev);
-
 	return rp->status;
 }
 
@@ -2312,10 +2308,8 @@ static void hci_cs_inquiry(struct hci_dev *hdev, __u8 status)
 {
 	bt_dev_dbg(hdev, "status 0x%2.2x", status);
 
-	if (status) {
-		hci_conn_check_pending(hdev);
+	if (status)
 		return;
-	}
 
 	if (hci_sent_cmd_data(hdev, HCI_OP_INQUIRY))
 		set_bit(HCI_INQUIRY, &hdev->flags);
@@ -2340,12 +2334,9 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status)
 
 	if (status) {
 		if (conn && conn->state == BT_CONNECT) {
-			if (status != HCI_ERROR_COMMAND_DISALLOWED || conn->attempt > 2) {
-				conn->state = BT_CLOSED;
-				hci_connect_cfm(conn, status);
-				hci_conn_del(conn);
-			} else
-				conn->state = BT_CONNECT2;
+			conn->state = BT_CLOSED;
+			hci_connect_cfm(conn, status);
+			hci_conn_del(conn);
 		}
 	} else {
 		if (!conn) {
@@ -3035,8 +3026,6 @@ static void hci_inquiry_complete_evt(struct hci_dev *hdev, void *data,
 
 	bt_dev_dbg(hdev, "status 0x%2.2x", ev->status);
 
-	hci_conn_check_pending(hdev);
-
 	if (!test_and_clear_bit(HCI_INQUIRY, &hdev->flags))
 		return;
 
@@ -3283,8 +3272,6 @@ static void hci_conn_complete_evt(struct hci_dev *hdev, void *data,
 
 unlock:
 	hci_dev_unlock(hdev);
-
-	hci_conn_check_pending(hdev);
 }
 
 static void hci_reject_conn(struct hci_dev *hdev, bdaddr_t *bdaddr)
Signature · 1 line
2.43.0
author maintainer
#80 lore raw

PATCH 80/228 Bluetooth: hci_conn: Always use sk_timeo as conn_timeout

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit bf98feea5b65ced367a871cf35fc044dedbcfb85 ]

This aligns the use socket sk_timeo as conn_timeout when initiating a connection and then use it when scheduling the resulting HCI command, that way the command is actually aborted synchronously thus not blocking commands generated by hci_abort_conn_sync to inform the controller the connection is to be aborted.

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci_core.h |  5 +++--
 include/net/bluetooth/l2cap.h    |  2 +-
 net/bluetooth/6lowpan.c          |  2 +-
 net/bluetooth/hci_conn.c         |  8 +++++---
 net/bluetooth/hci_sync.c         |  2 +-
 net/bluetooth/l2cap_core.c       | 10 ++++------
 net/bluetooth/l2cap_sock.c       |  3 ++-
 net/bluetooth/mgmt.c             |  3 ++-
 net/bluetooth/sco.c              |  3 ++-
 9 files changed, 21 insertions(+), 17 deletions(-)
include/net/bluetooth/hci_core.h +3 -2
index ddbf287ee64fb..65da50f216069 100644
@@ -1496,9 +1496,10 @@ struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst,
 				u16 conn_timeout, u8 role);
 struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 				 u8 sec_level, u8 auth_type,
-				 enum conn_reasons conn_reason);
+				 enum conn_reasons conn_reason, u16 timeout);
 struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst,
-				 __u16 setting, struct bt_codec *codec);
+				 __u16 setting, struct bt_codec *codec,
+				 u16 timeout);
 struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst,
 			      __u8 dst_type, struct bt_iso_qos *qos);
 struct hci_conn *hci_bind_bis(struct hci_dev *hdev, bdaddr_t *dst,
include/net/bluetooth/l2cap.h +1 -1
index 92d7197f9a563..a4278aa618ab1 100644
@@ -939,7 +939,7 @@ int l2cap_add_scid(struct l2cap_chan *chan,  __u16 scid);
 struct l2cap_chan *l2cap_chan_create(void);
 void l2cap_chan_close(struct l2cap_chan *chan, int reason);
 int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid,
-		       bdaddr_t *dst, u8 dst_type);
+		       bdaddr_t *dst, u8 dst_type, u16 timeout);
 int l2cap_chan_reconfigure(struct l2cap_chan *chan, __u16 mtu);
 int l2cap_chan_send(struct l2cap_chan *chan, struct msghdr *msg, size_t len);
 void l2cap_chan_busy(struct l2cap_chan *chan, int busy);
net/bluetooth/6lowpan.c +1 -1
index 4eb1b3ced0d27..715cbafbf6631 100644
@@ -892,7 +892,7 @@ static int bt_6lowpan_connect(bdaddr_t *addr, u8 dst_type)
 	chan->ops = &bt_6lowpan_chan_ops;
 
 	err = l2cap_chan_connect(chan, cpu_to_le16(L2CAP_PSM_IPSP), 0,
-				 addr, dst_type);
+				 addr, dst_type, L2CAP_CONN_TIMEOUT);
 
 	BT_DBG("chan %p err %d", chan, err);
 	if (err < 0)
net/bluetooth/hci_conn.c +5 -3
index 13481f08e47ee..47f117874479d 100644
@@ -1607,7 +1607,7 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst,
 
 struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 				 u8 sec_level, u8 auth_type,
-				 enum conn_reasons conn_reason)
+				 enum conn_reasons conn_reason, u16 timeout)
 {
 	struct hci_conn *acl;
 
@@ -1643,6 +1643,7 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 		acl->sec_level = BT_SECURITY_LOW;
 		acl->pending_sec_level = sec_level;
 		acl->auth_type = auth_type;
+		acl->conn_timeout = timeout;
 
 		err = hci_acl_create_connection_sync(hdev, acl);
 		if (err) {
@@ -1683,14 +1684,15 @@ static struct hci_link *hci_conn_link(struct hci_conn *parent,
 }
 
 struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst,
-				 __u16 setting, struct bt_codec *codec)
+				 __u16 setting, struct bt_codec *codec,
+				 u16 timeout)
 {
 	struct hci_conn *acl;
 	struct hci_conn *sco;
 	struct hci_link *link;
 
 	acl = hci_connect_acl(hdev, dst, BT_SECURITY_LOW, HCI_AT_NO_BONDING,
-			      CONN_REASON_SCO_CONNECT);
+			      CONN_REASON_SCO_CONNECT, timeout);
 	if (IS_ERR(acl))
 		return acl;
 
net/bluetooth/hci_sync.c +1 -1
index e4f89b6e917e2..816fd9c38ae04 100644
@@ -6607,7 +6607,7 @@ static int __hci_acl_create_connection_sync(struct hci_dev *hdev, void *data)
 	err = __hci_cmd_sync_status_sk(hdev, HCI_OP_CREATE_CONN,
 				       sizeof(cp), &cp,
 				       HCI_EV_CONN_COMPLETE,
-				       HCI_ACL_CONN_TIMEOUT, NULL);
+				       conn->conn_timeout, NULL);
 
 	if (err == -ETIMEDOUT)
 		hci_abort_conn_sync(hdev, conn, HCI_ERROR_LOCAL_HOST_TERM);
net/bluetooth/l2cap_core.c +4 -6
index 706d2478ddb33..dc08974087936 100644
@@ -6924,7 +6924,7 @@ static void l2cap_chan_by_pid(struct l2cap_chan *chan, void *data)
 }
 
 int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid,
-		       bdaddr_t *dst, u8 dst_type)
+		       bdaddr_t *dst, u8 dst_type, u16 timeout)
 {
 	struct l2cap_conn *conn;
 	struct hci_conn *hcon;
@@ -7017,19 +7017,17 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid,
 
 		if (hci_dev_test_flag(hdev, HCI_ADVERTISING))
 			hcon = hci_connect_le(hdev, dst, dst_type, false,
-					      chan->sec_level,
-					      HCI_LE_CONN_TIMEOUT,
+					      chan->sec_level, timeout,
 					      HCI_ROLE_SLAVE);
 		else
 			hcon = hci_connect_le_scan(hdev, dst, dst_type,
-						   chan->sec_level,
-						   HCI_LE_CONN_TIMEOUT,
+						   chan->sec_level, timeout,
 						   CONN_REASON_L2CAP_CHAN);
 
 	} else {
 		u8 auth_type = l2cap_get_auth_type(chan);
 		hcon = hci_connect_acl(hdev, dst, chan->sec_level, auth_type,
-				       CONN_REASON_L2CAP_CHAN);
+				       CONN_REASON_L2CAP_CHAN, timeout);
 	}
 
 	if (IS_ERR(hcon)) {
net/bluetooth/l2cap_sock.c +2 -1
index 1eeea5d1306c2..e7d810b23082f 100644
@@ -254,7 +254,8 @@ static int l2cap_sock_connect(struct socket *sock, struct sockaddr *addr,
 		chan->mode = L2CAP_MODE_LE_FLOWCTL;
 
 	err = l2cap_chan_connect(chan, la.l2_psm, __le16_to_cpu(la.l2_cid),
-				 &la.l2_bdaddr, la.l2_bdaddr_type);
+				 &la.l2_bdaddr, la.l2_bdaddr_type,
+				 sk->sk_sndtimeo);
 	if (err)
 		return err;
 
net/bluetooth/mgmt.c +2 -1
index 640d6d54ac6ba..0759975be6479 100644
@@ -3436,7 +3436,8 @@ static int pair_device(struct sock *sk, struct hci_dev *hdev, void *data,
 
 	if (cp->addr.type == BDADDR_BREDR) {
 		conn = hci_connect_acl(hdev, &cp->addr.bdaddr, sec_level,
-				       auth_type, CONN_REASON_PAIR_DEVICE);
+				       auth_type, CONN_REASON_PAIR_DEVICE,
+				       HCI_ACL_CONN_TIMEOUT);
 	} else {
 		u8 addr_type = le_addr_type(cp->addr.type);
 		struct hci_conn_params *p;
net/bluetooth/sco.c +2 -1
index 8e4f39b8601cb..368e026f4d15c 100644
@@ -264,7 +264,8 @@ static int sco_connect(struct sock *sk)
 	}
 
 	hcon = hci_connect_sco(hdev, type, &sco_pi(sk)->dst,
-			       sco_pi(sk)->setting, &sco_pi(sk)->codec);
+			       sco_pi(sk)->setting, &sco_pi(sk)->codec,
+			       sk->sk_sndtimeo);
 	if (IS_ERR(hcon)) {
 		err = PTR_ERR(hcon);
 		goto unlock;
Signature · 1 line
2.43.0
author maintainer
#81 lore raw

PATCH 81/228 Bluetooth: hci_conn: Fix UAF Write in __hci_acl_create_connection_sync

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit 5f641f03abccddd1a37233ff1b8e774b9ff1f4e8 ]

This fixes the UAF on __hci_acl_create_connection_sync caused by connection abortion, it uses the same logic as to LE_LINK which uses hci_cmd_sync_cancel to prevent the callback to run if the connection is abort prematurely.

Reported-by [email protected]
Fixes 45340097ce6e ("Bluetooth: hci_conn: Only do ACL connections sequentially")

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci_sync.h |  3 +--
 net/bluetooth/hci_conn.c         |  3 ++-
 net/bluetooth/hci_sync.c         | 16 ++++++++++------
 3 files changed, 13 insertions(+), 9 deletions(-)
include/net/bluetooth/hci_sync.h +1 -2
index 824660f8f30da..ed334c253ebcd 100644
@@ -139,5 +139,4 @@ int hci_le_big_terminate_sync(struct hci_dev *hdev, u8 handle);
 
 int hci_le_pa_terminate_sync(struct hci_dev *hdev, u16 handle);
 
-int hci_acl_create_connection_sync(struct hci_dev *hdev,
-				   struct hci_conn *conn);
+int hci_connect_acl_sync(struct hci_dev *hdev, struct hci_conn *conn);
net/bluetooth/hci_conn.c +2 -1
index 47f117874479d..3475ea55c2534 100644
@@ -1645,7 +1645,7 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 		acl->auth_type = auth_type;
 		acl->conn_timeout = timeout;
 
-		err = hci_acl_create_connection_sync(hdev, acl);
+		err = hci_connect_acl_sync(hdev, acl);
 		if (err) {
 			hci_conn_del(acl);
 			return ERR_PTR(err);
@@ -2942,6 +2942,7 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason)
 	 */
 	if (conn->state == BT_CONNECT && hdev->req_status == HCI_REQ_PEND) {
 		switch (hci_skb_event(hdev->sent_cmd)) {
+		case HCI_EV_CONN_COMPLETE:
 		case HCI_EV_LE_CONN_COMPLETE:
 		case HCI_EV_LE_ENHANCED_CONN_COMPLETE:
 		case HCI_EVT_LE_CIS_ESTABLISHED:
net/bluetooth/hci_sync.c +10 -6
index 816fd9c38ae04..cffa61ecf234f 100644
@@ -6552,13 +6552,18 @@ int hci_update_adv_data(struct hci_dev *hdev, u8 instance)
 				  UINT_PTR(instance), NULL);
 }
 
-static int __hci_acl_create_connection_sync(struct hci_dev *hdev, void *data)
+static int hci_acl_create_conn_sync(struct hci_dev *hdev, void *data)
 {
-	struct hci_conn *conn = data;
+	struct hci_conn *conn;
+	u16 handle = PTR_UINT(data);
 	struct inquiry_entry *ie;
 	struct hci_cp_create_conn cp;
 	int err;
 
+	conn = hci_conn_hash_lookup_handle(hdev, handle);
+	if (!conn)
+		return 0;
+
 	/* Many controllers disallow HCI Create Connection while it is doing
 	 * HCI Inquiry. So we cancel the Inquiry first before issuing HCI Create
 	 * Connection. This may cause the MGMT discovering state to become false
@@ -6615,9 +6620,8 @@ static int __hci_acl_create_connection_sync(struct hci_dev *hdev, void *data)
 	return err;
 }
 
-int hci_acl_create_connection_sync(struct hci_dev *hdev,
-				   struct hci_conn *conn)
+int hci_connect_acl_sync(struct hci_dev *hdev, struct hci_conn *conn)
 {
-	return hci_cmd_sync_queue(hdev, __hci_acl_create_connection_sync,
-				  conn, NULL);
+	return hci_cmd_sync_queue(hdev, hci_acl_create_conn_sync,
+				  UINT_PTR(conn->handle), NULL);
 }
Signature · 1 line
2.43.0
author maintainer
#82 lore raw

PATCH 82/228 Bluetooth: hci_sync: Add helper functions to manipulate cmd_sync queue

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit 505ea2b295929e7be2b4e1bc86ee31cb7862fb01 ]

This adds functions to queue, dequeue and lookup into the cmd_sync list.

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci_sync.h |  12 +++
 net/bluetooth/hci_sync.c         | 132 +++++++++++++++++++++++++++++--
 2 files changed, 136 insertions(+), 8 deletions(-)
include/net/bluetooth/hci_sync.h +12 -0
index ed334c253ebcd..4ff4aa68ee196 100644
@@ -48,6 +48,18 @@ int hci_cmd_sync_submit(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
 			void *data, hci_cmd_sync_work_destroy_t destroy);
 int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
 		       void *data, hci_cmd_sync_work_destroy_t destroy);
+struct hci_cmd_sync_work_entry *
+hci_cmd_sync_lookup_entry(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
+			  void *data, hci_cmd_sync_work_destroy_t destroy);
+int hci_cmd_sync_queue_once(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
+			    void *data, hci_cmd_sync_work_destroy_t destroy);
+void hci_cmd_sync_cancel_entry(struct hci_dev *hdev,
+			       struct hci_cmd_sync_work_entry *entry);
+bool hci_cmd_sync_dequeue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
+			  void *data, hci_cmd_sync_work_destroy_t destroy);
+bool hci_cmd_sync_dequeue_once(struct hci_dev *hdev,
+			      hci_cmd_sync_work_func_t func, void *data,
+			      hci_cmd_sync_work_destroy_t destroy);
 
 int hci_update_eir_sync(struct hci_dev *hdev);
 int hci_update_class_sync(struct hci_dev *hdev);
net/bluetooth/hci_sync.c +124 -8

156 lines not on this page · raw message

Signature · 1 line
2.43.0
author maintainer
#83 lore raw

PATCH 83/228 Bluetooth: hci_sync: Attempt to dequeue connection attempt

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit 881559af5f5c545f6828e7c74d79813eb886d523 ]

If connection is still queued/pending in the cmd_sync queue it means no command has been generated and it should be safe to just dequeue the callback when it is being aborted.

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci_core.h | 19 ++++++++
 include/net/bluetooth/hci_sync.h | 10 +++--
 net/bluetooth/hci_conn.c         | 70 ++++++------------------------
 net/bluetooth/hci_sync.c         | 74 ++++++++++++++++++++++++++++----
 4 files changed, 102 insertions(+), 71 deletions(-)
include/net/bluetooth/hci_core.h +19 -0
index 65da50f216069..93e2b17b11267 100644
@@ -1084,6 +1084,24 @@ static inline unsigned int hci_conn_count(struct hci_dev *hdev)
 	return c->acl_num + c->amp_num + c->sco_num + c->le_num + c->iso_num;
 }
 
+static inline bool hci_conn_valid(struct hci_dev *hdev, struct hci_conn *conn)
+{
+	struct hci_conn_hash *h = &hdev->conn_hash;
+	struct hci_conn  *c;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(c, &h->list, list) {
+		if (c == conn) {
+			rcu_read_unlock();
+			return true;
+		}
+	}
+	rcu_read_unlock();
+
+	return false;
+}
+
 static inline __u8 hci_conn_lookup_type(struct hci_dev *hdev, __u16 handle)
 {
 	struct hci_conn_hash *h = &hdev->conn_hash;
@@ -1494,6 +1512,7 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst,
 struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst,
 				u8 dst_type, bool dst_resolved, u8 sec_level,
 				u16 conn_timeout, u8 role);
+void hci_connect_le_scan_cleanup(struct hci_conn *conn, u8 status);
 struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 				 u8 sec_level, u8 auth_type,
 				 enum conn_reasons conn_reason, u16 timeout);
include/net/bluetooth/hci_sync.h +6 -4
index 4ff4aa68ee196..6a9d063e9f472 100644
@@ -48,11 +48,11 @@ int hci_cmd_sync_submit(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
 			void *data, hci_cmd_sync_work_destroy_t destroy);
 int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
 		       void *data, hci_cmd_sync_work_destroy_t destroy);
+int hci_cmd_sync_queue_once(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
+			    void *data, hci_cmd_sync_work_destroy_t destroy);
 struct hci_cmd_sync_work_entry *
 hci_cmd_sync_lookup_entry(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
 			  void *data, hci_cmd_sync_work_destroy_t destroy);
-int hci_cmd_sync_queue_once(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
-			    void *data, hci_cmd_sync_work_destroy_t destroy);
 void hci_cmd_sync_cancel_entry(struct hci_dev *hdev,
 			       struct hci_cmd_sync_work_entry *entry);
 bool hci_cmd_sync_dequeue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
@@ -139,8 +139,6 @@ struct hci_conn;
 
 int hci_abort_conn_sync(struct hci_dev *hdev, struct hci_conn *conn, u8 reason);
 
-int hci_le_create_conn_sync(struct hci_dev *hdev, struct hci_conn *conn);
-
 int hci_le_create_cis_sync(struct hci_dev *hdev);
 
 int hci_le_remove_cig_sync(struct hci_dev *hdev, u8 handle);
@@ -152,3 +150,7 @@ int hci_le_big_terminate_sync(struct hci_dev *hdev, u8 handle);
 int hci_le_pa_terminate_sync(struct hci_dev *hdev, u16 handle);
 
 int hci_connect_acl_sync(struct hci_dev *hdev, struct hci_conn *conn);
+
+int hci_connect_le_sync(struct hci_dev *hdev, struct hci_conn *conn);
+
+int hci_cancel_connect_sync(struct hci_dev *hdev, struct hci_conn *conn);
net/bluetooth/hci_conn.c +12 -58

107 lines not on this page · raw message

net/bluetooth/hci_sync.c +65 -9

99 lines not on this page · raw message

Signature · 1 line
2.43.0
author maintainer
#84 lore raw

PATCH 84/228 Bluetooth: ISO: Reassemble PA data for bcast sink

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Iulia Tanasescu <[email protected]>

[ Upstream commit 168d9bf9c7f01df71e6404cfff66d9c2a8e968fb ]

This adds support to reassemble PA data for a Broadcast Sink listening socket. This is needed in case the BASE is received fragmented in multiple PA reports.

PA data is first reassembled inside the hcon, before the BASE is extracted and stored inside the socket. The length of the le_per_adv_data hcon array has been raised to 1650, to accommodate the maximum PA data length that can come fragmented, according to spec.

Stable-dep-of: 2e7ed5f5e69b ("Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync")

---
 include/net/bluetooth/hci.h      |  5 ++++
 include/net/bluetooth/hci_core.h |  5 ++--
 net/bluetooth/iso.c              | 50 +++++++++++++++++++++++++++++---
 3 files changed, 54 insertions(+), 6 deletions(-)
include/net/bluetooth/hci.h +5 -0
index bfddb648fcc3e..2cfd8d862639f 100644
@@ -2053,6 +2053,7 @@ struct hci_cp_le_set_per_adv_params {
 } __packed;
 
 #define HCI_MAX_PER_AD_LENGTH	252
+#define HCI_MAX_PER_AD_TOT_LEN	1650
 
 #define HCI_OP_LE_SET_PER_ADV_DATA		0x203f
 struct hci_cp_le_set_per_adv_data {
@@ -2813,6 +2814,10 @@ struct hci_ev_le_per_adv_report {
 	__u8     data[];
 } __packed;
 
+#define LE_PA_DATA_COMPLETE	0x00
+#define LE_PA_DATA_MORE_TO_COME	0x01
+#define LE_PA_DATA_TRUNCATED	0x02
+
 #define HCI_EV_LE_EXT_ADV_SET_TERM	0x12
 struct hci_evt_le_ext_adv_set_term {
 	__u8	status;
include/net/bluetooth/hci_core.h +3 -2
index 93e2b17b11267..1da71254caf61 100644
@@ -735,8 +735,9 @@ struct hci_conn {
 	__u16		le_supv_timeout;
 	__u8		le_adv_data[HCI_MAX_EXT_AD_LENGTH];
 	__u8		le_adv_data_len;
-	__u8		le_per_adv_data[HCI_MAX_PER_AD_LENGTH];
-	__u8		le_per_adv_data_len;
+	__u8		le_per_adv_data[HCI_MAX_PER_AD_TOT_LEN];
+	__u16		le_per_adv_data_len;
+	__u16		le_per_adv_data_offset;
 	__u8		le_tx_phy;
 	__u8		le_rx_phy;
 	__s8		rssi;
net/bluetooth/iso.c +46 -4

62 lines not on this page · raw message

Signature · 1 line
2.43.0
author maintainer
#85 lore raw

PATCH 85/228 Bluetooth: hci_sync: Use advertised PHYs on hci_le_ext_create_conn_sync

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit 2e7ed5f5e69b6fe93dd3c6b651d041e0a7a456d1 ]

The extended advertising reports do report the PHYs so this store then in hci_conn so it can be later used in hci_le_ext_create_conn_sync to narrow the PHYs to be scanned since the controller will also perform a scan having a smaller set of PHYs shall reduce the time it takes to find and connect peers.

Fixes 288c90224eec ("Bluetooth: Enable all supported LE PHY by default")
---
 include/net/bluetooth/hci_core.h |  4 +++-
 net/bluetooth/hci_conn.c         |  6 ++++--
 net/bluetooth/hci_event.c        | 20 ++++++++++++--------
 net/bluetooth/hci_sync.c         |  9 ++++++---
 net/bluetooth/l2cap_core.c       |  2 +-
 5 files changed, 26 insertions(+), 15 deletions(-)
include/net/bluetooth/hci_core.h +3 -1
index 1da71254caf61..b6fb104ea1976 100644
@@ -738,6 +738,8 @@ struct hci_conn {
 	__u8		le_per_adv_data[HCI_MAX_PER_AD_TOT_LEN];
 	__u16		le_per_adv_data_len;
 	__u16		le_per_adv_data_offset;
+	__u8		le_adv_phy;
+	__u8		le_adv_sec_phy;
 	__u8		le_tx_phy;
 	__u8		le_rx_phy;
 	__s8		rssi;
@@ -1512,7 +1514,7 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst,
 				     enum conn_reasons conn_reason);
 struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst,
 				u8 dst_type, bool dst_resolved, u8 sec_level,
-				u16 conn_timeout, u8 role);
+				u16 conn_timeout, u8 role, u8 phy, u8 sec_phy);
 void hci_connect_le_scan_cleanup(struct hci_conn *conn, u8 status);
 struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 				 u8 sec_level, u8 auth_type,
net/bluetooth/hci_conn.c +4 -2
index 3a66d357b9323..18f97b2288699 100644
@@ -1263,7 +1263,7 @@ u8 hci_conn_set_handle(struct hci_conn *conn, u16 handle)
 
 struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst,
 				u8 dst_type, bool dst_resolved, u8 sec_level,
-				u16 conn_timeout, u8 role)
+				u16 conn_timeout, u8 role, u8 phy, u8 sec_phy)
 {
 	struct hci_conn *conn;
 	struct smp_irk *irk;
@@ -1326,6 +1326,8 @@ struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst,
 	conn->dst_type = dst_type;
 	conn->sec_level = BT_SECURITY_LOW;
 	conn->conn_timeout = conn_timeout;
+	conn->le_adv_phy = phy;
+	conn->le_adv_sec_phy = sec_phy;
 
 	err = hci_connect_le_sync(hdev, conn);
 	if (err) {
@@ -2253,7 +2255,7 @@ struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst,
 		le = hci_connect_le(hdev, dst, dst_type, false,
 				    BT_SECURITY_LOW,
 				    HCI_LE_CONN_TIMEOUT,
-				    HCI_ROLE_SLAVE);
+				    HCI_ROLE_SLAVE, 0, 0);
 	else
 		le = hci_connect_le_scan(hdev, dst, dst_type,
 					 BT_SECURITY_LOW,
net/bluetooth/hci_event.c +12 -8
index f5d77c8089d81..9450df4ee5b9b 100644
@@ -6182,7 +6182,7 @@ static void hci_le_conn_update_complete_evt(struct hci_dev *hdev, void *data,
 static struct hci_conn *check_pending_le_conn(struct hci_dev *hdev,
 					      bdaddr_t *addr,
 					      u8 addr_type, bool addr_resolved,
-					      u8 adv_type)
+					      u8 adv_type, u8 phy, u8 sec_phy)
 {
 	struct hci_conn *conn;
 	struct hci_conn_params *params;
@@ -6237,7 +6237,7 @@ static struct hci_conn *check_pending_le_conn(struct hci_dev *hdev,
 
 	conn = hci_connect_le(hdev, addr, addr_type, addr_resolved,
 			      BT_SECURITY_LOW, hdev->def_le_autoconnect_timeout,
-			      HCI_ROLE_MASTER);
+			      HCI_ROLE_MASTER, phy, sec_phy);
 	if (!IS_ERR(conn)) {
 		/* If HCI_AUTO_CONN_EXPLICIT is set, conn is already owned
 		 * by higher layer that tried to connect, if no then
@@ -6272,8 +6272,9 @@ static struct hci_conn *check_pending_le_conn(struct hci_dev *hdev,
 
 static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr,
 			       u8 bdaddr_type, bdaddr_t *direct_addr,
-			       u8 direct_addr_type, s8 rssi, u8 *data, u8 len,
-			       bool ext_adv, bool ctl_time, u64 instant)
+			       u8 direct_addr_type, u8 phy, u8 sec_phy, s8 rssi,
+			       u8 *data, u8 len, bool ext_adv, bool ctl_time,
+			       u64 instant)
 {
 	struct discovery_state *d = &hdev->discovery;
 	struct smp_irk *irk;
@@ -6361,7 +6362,7 @@ static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr,
 	 * for advertising reports) and is already verified to be RPA above.
 	 */
 	conn = check_pending_le_conn(hdev, bdaddr, bdaddr_type, bdaddr_resolved,
-				     type);
+				     type, phy, sec_phy);
 	if (!ext_adv && conn && type == LE_ADV_IND &&
 	    len <= max_adv_len(hdev)) {
 		/* Store report for later inclusion by
@@ -6507,7 +6508,8 @@ static void hci_le_adv_report_evt(struct hci_dev *hdev, void *data,
 		if (info->length <= max_adv_len(hdev)) {
 			rssi = info->data[info->length];
 			process_adv_report(hdev, info->type, &info->bdaddr,
-					   info->bdaddr_type, NULL, 0, rssi,
+					   info->bdaddr_type, NULL, 0,
+					   HCI_ADV_PHY_1M, 0, rssi,
 					   info->data, info->length, false,
 					   false, instant);
 		} else {
@@ -6592,6 +6594,8 @@ static void hci_le_ext_adv_report_evt(struct hci_dev *hdev, void *data,
 		if (legacy_evt_type != LE_ADV_INVALID) {
 			process_adv_report(hdev, legacy_evt_type, &info->bdaddr,
 					   info->bdaddr_type, NULL, 0,
+					   info->primary_phy,
+					   info->secondary_phy,
 					   info->rssi, info->data, info->length,
 					   !(evt_type & LE_EXT_ADV_LEGACY_PDU),
 					   false, instant);
@@ -6874,8 +6878,8 @@ static void hci_le_direct_adv_report_evt(struct hci_dev *hdev, void *data,
 
 		process_adv_report(hdev, info->type, &info->bdaddr,
 				   info->bdaddr_type, &info->direct_addr,
-				   info->direct_addr_type, info->rssi, NULL, 0,
-				   false, false, instant);
+				   info->direct_addr_type, HCI_ADV_PHY_1M, 0,
+				   info->rssi, NULL, 0, false, false, instant);
 	}
 
 	hci_dev_unlock(hdev);
net/bluetooth/hci_sync.c +6 -3
index ed7db30ac0e0d..84fe44c796f4f 100644
@@ -6315,7 +6315,8 @@ static int hci_le_ext_create_conn_sync(struct hci_dev *hdev,
 
 	plen = sizeof(*cp);
 
-	if (scan_1m(hdev)) {
+	if (scan_1m(hdev) && (conn->le_adv_phy == HCI_ADV_PHY_1M ||
+			      conn->le_adv_sec_phy == HCI_ADV_PHY_1M)) {
 		cp->phys |= LE_SCAN_PHY_1M;
 		set_ext_conn_params(conn, p);
 
@@ -6323,7 +6324,8 @@ static int hci_le_ext_create_conn_sync(struct hci_dev *hdev,
 		plen += sizeof(*p);
 	}
 
-	if (scan_2m(hdev)) {
+	if (scan_2m(hdev) && (conn->le_adv_phy == HCI_ADV_PHY_2M ||
+			      conn->le_adv_sec_phy == HCI_ADV_PHY_2M)) {
 		cp->phys |= LE_SCAN_PHY_2M;
 		set_ext_conn_params(conn, p);
 
@@ -6331,7 +6333,8 @@ static int hci_le_ext_create_conn_sync(struct hci_dev *hdev,
 		plen += sizeof(*p);
 	}
 
-	if (scan_coded(hdev)) {
+	if (scan_coded(hdev) && (conn->le_adv_phy == HCI_ADV_PHY_CODED ||
+				 conn->le_adv_sec_phy == HCI_ADV_PHY_CODED)) {
 		cp->phys |= LE_SCAN_PHY_CODED;
 		set_ext_conn_params(conn, p);
 
net/bluetooth/l2cap_core.c +1 -1
index dc08974087936..84fc70862d78a 100644
@@ -7018,7 +7018,7 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid,
 		if (hci_dev_test_flag(hdev, HCI_ADVERTISING))
 			hcon = hci_connect_le(hdev, dst, dst_type, false,
 					      chan->sec_level, timeout,
-					      HCI_ROLE_SLAVE);
+					      HCI_ROLE_SLAVE, 0, 0);
 		else
 			hcon = hci_connect_le_scan(hdev, dst, dst_type,
 						   chan->sec_level, timeout,
Signature · 1 line
2.43.0
author maintainer
#86 lore raw

PATCH 86/228 Bluetooth: btusb: Fix triggering coredump implementation for QCA

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zijun Hu <[email protected]>

[ Upstream commit b23d98d46d2858dcc0fd016caff165cbdc24e70a ]

btusb_coredump_qca() uses __hci_cmd_sync() to send a vendor-specific command to trigger firmware coredump, but the command does not have any event as its sync response, so it is not suitable to use __hci_cmd_sync(), fixed by using __hci_cmd_send().

Fixes 20981ce2d5a5 ("Bluetooth: btusb: Add WCN6855 devcoredump support")
---
 drivers/bluetooth/btusb.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
drivers/bluetooth/btusb.c +4 -5
index f684108dc2f1c..58faada874f5c 100644
@@ -3463,13 +3463,12 @@ static void btusb_dump_hdr_qca(struct hci_dev *hdev, struct sk_buff *skb)
 
 static void btusb_coredump_qca(struct hci_dev *hdev)
 {
+	int err;
 	static const u8 param[] = { 0x26 };
-	struct sk_buff *skb;
 
-	skb = __hci_cmd_sync(hdev, 0xfc0c, 1, param, HCI_CMD_TIMEOUT);
-	if (IS_ERR(skb))
-		bt_dev_err(hdev, "%s: triggle crash failed (%ld)", __func__, PTR_ERR(skb));
-	kfree_skb(skb);
+	err = __hci_cmd_send(hdev, 0xfc0c, 1, param);
+	if (err < 0)
+		bt_dev_err(hdev, "%s: triggle crash failed (%d)", __func__, err);
 }
 
 /*
Signature · 1 line
2.43.0
author maintainer
#87 lore raw

PATCH 87/228 Bluetooth: hci_event: Fix sending HCI_OP_READ_ENC_KEY_SIZE

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit a9a830a676a9a93c5020f5c61236166931fa4266 ]

The code shall always check if HCI_QUIRK_BROKEN_READ_ENC_KEY_SIZE has been set before attempting to use HCI_OP_READ_ENC_KEY_SIZE.

Fixes c569242cd492 ("Bluetooth: hci_event: set the conn encrypted before conn establishes")
---
 include/net/bluetooth/hci_core.h | 4 ++++
 net/bluetooth/hci_event.c        | 5 ++---
 2 files changed, 6 insertions(+), 3 deletions(-)
include/net/bluetooth/hci_core.h +4 -0
index b6fb104ea1976..fe9e1524d30ff 100644
@@ -1907,6 +1907,10 @@ void hci_conn_del_sysfs(struct hci_conn *conn);
 #define privacy_mode_capable(dev) (use_ll_privacy(dev) && \
 				   (hdev->commands[39] & 0x04))
 
+#define read_key_size_capable(dev) \
+	((dev)->commands[20] & 0x10 && \
+	 !test_bit(HCI_QUIRK_BROKEN_READ_ENC_KEY_SIZE, &hdev->quirks))
+
 /* Use enhanced synchronous connection if command is supported and its quirk
  * has not been set.
  */
net/bluetooth/hci_event.c +2 -3
index 9450df4ee5b9b..9d1063c51ed29 100644
@@ -3218,7 +3218,7 @@ static void hci_conn_complete_evt(struct hci_dev *hdev, void *data,
 			if (key) {
 				set_bit(HCI_CONN_ENCRYPT, &conn->flags);
 
-				if (!(hdev->commands[20] & 0x10)) {
+				if (!read_key_size_capable(hdev)) {
 					conn->enc_key_size = HCI_LINK_KEY_SIZE;
 				} else {
 					cp.handle = cpu_to_le16(conn->handle);
@@ -3666,8 +3666,7 @@ static void hci_encrypt_change_evt(struct hci_dev *hdev, void *data,
 		 * controller really supports it. If it doesn't, assume
 		 * the default size (16).
 		 */
-		if (!(hdev->commands[20] & 0x10) ||
-		    test_bit(HCI_QUIRK_BROKEN_READ_ENC_KEY_SIZE, &hdev->quirks)) {
+		if (!read_key_size_capable(hdev)) {
 			conn->enc_key_size = HCI_LINK_KEY_SIZE;
 			goto notify;
 		}
Signature · 1 line
2.43.0
author maintainer
#88 lore raw

PATCH 88/228 Bluetooth: MGMT: Fix failing to MGMT_OP_ADD_UUID/MGMT_OP_REMOVE_UUID

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit 6eb5fcc416f127f220b9177a5c9ae751cac1cda8 ]

These commands don't require the adapter to be up and running so don't use hci_cmd_sync_queue which would check that flag, instead use hci_cmd_sync_submit which would ensure mgmt_class_complete is set properly regardless if any command was actually run or not.

Fixes d883a4669a1d ("Bluetooth: hci_sync: Only allow hci_cmd_sync_queue if running")
---
 net/bluetooth/mgmt.c | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)
net/bluetooth/mgmt.c +15 -5
index 0759975be6479..7e56cb3d377a1 100644
@@ -2616,7 +2616,11 @@ static int add_uuid(struct sock *sk, struct hci_dev *hdev, void *data, u16 len)
 		goto failed;
 	}
 
-	err = hci_cmd_sync_queue(hdev, add_uuid_sync, cmd, mgmt_class_complete);
+	/* MGMT_OP_ADD_UUID don't require adapter the UP/Running so use
+	 * hci_cmd_sync_submit instead of hci_cmd_sync_queue.
+	 */
+	err = hci_cmd_sync_submit(hdev, add_uuid_sync, cmd,
+				  mgmt_class_complete);
 	if (err < 0) {
 		mgmt_pending_free(cmd);
 		goto failed;
@@ -2710,8 +2714,11 @@ static int remove_uuid(struct sock *sk, struct hci_dev *hdev, void *data,
 		goto unlock;
 	}
 
-	err = hci_cmd_sync_queue(hdev, remove_uuid_sync, cmd,
-				 mgmt_class_complete);
+	/* MGMT_OP_REMOVE_UUID don't require adapter the UP/Running so use
+	 * hci_cmd_sync_submit instead of hci_cmd_sync_queue.
+	 */
+	err = hci_cmd_sync_submit(hdev, remove_uuid_sync, cmd,
+				  mgmt_class_complete);
 	if (err < 0)
 		mgmt_pending_free(cmd);
 
@@ -2777,8 +2784,11 @@ static int set_dev_class(struct sock *sk, struct hci_dev *hdev, void *data,
 		goto unlock;
 	}
 
-	err = hci_cmd_sync_queue(hdev, set_class_sync, cmd,
-				 mgmt_class_complete);
+	/* MGMT_OP_SET_DEV_CLASS don't require adapter the UP/Running so use
+	 * hci_cmd_sync_submit instead of hci_cmd_sync_queue.
+	 */
+	err = hci_cmd_sync_submit(hdev, set_class_sync, cmd,
+				  mgmt_class_complete);
 	if (err < 0)
 		mgmt_pending_free(cmd);
 
Signature · 1 line
2.43.0
author maintainer
#89 lore raw

PATCH 89/228 Bluetooth: btusb: mediatek: Fix double free of skb in coredump

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Wang <[email protected]>

[ Upstream commit 18bdb386a1a30e7a3d7732a98e45e69cf6b5710d ]

hci_devcd_append() would free the skb on error so the caller don't have to free it again otherwise it would cause the double free of skb.

Fixes 0b7015132878 ("Bluetooth: btusb: mediatek: add MediaTek devcoredump support")
---
 drivers/bluetooth/btmtk.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
drivers/bluetooth/btmtk.c +3 -4
index ac8ebccd35075..812fd2a8f853e 100644
@@ -380,8 +380,10 @@ int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
 	switch (data->cd_info.state) {
 	case HCI_DEVCOREDUMP_IDLE:
 		err = hci_devcd_init(hdev, MTK_COREDUMP_SIZE);
-		if (err < 0)
+		if (err < 0) {
+			kfree_skb(skb);
 			break;
+		}
 		data->cd_info.cnt = 0;
 
 		/* It is supposed coredump can be done within 5 seconds */
@@ -407,9 +409,6 @@ int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
 		break;
 	}
 
-	if (err < 0)
-		kfree_skb(skb);
-
 	return err;
 }
 EXPORT_SYMBOL_GPL(btmtk_process_coredump);
Signature · 1 line
2.43.0
author maintainer
#90 lore raw

PATCH 90/228 Bluetooth: hci_sync: Using hci_cmd_sync_submit when removing Adv Monitor

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chun-Yi Lee <[email protected]>

[ Upstream commit 88cd6e6b2d327faa13e4505b07f1e380e51b21ff ]

Since the d883a4669a1de be introduced in v6.4, bluetooth daemon got the following failed message of MGMT_OP_REMOVE_ADV_MONITOR command when controller is power-off:

bluetoothd[20976]:
src/adapter.c:reset_adv_monitors_complete() Failed to reset Adv
Monitors: Failed>

Normally this situation is happened when the bluetoothd deamon be started manually after system booting. Which means that bluetoothd received MGMT_EV_INDEX_ADDED event after kernel runs hci_power_off().

Base on doc/mgmt-api.txt, the MGMT_OP_REMOVE_ADV_MONITOR command can be used when the controller is not powered. This patch changes the code in remove_adv_monitor() to use hci_cmd_sync_submit() instead of hci_cmd_sync_queue().

Fixes d883a4669a1de ("Bluetooth: hci_sync: Only allow hci_cmd_sync_queue if running")
Cc Luiz Augusto von Dentz <[email protected]>
Cc Manish Mandlik <[email protected]>
Cc Archie Pusaka <[email protected]>
Cc Miao-chen Chou <[email protected]>
---
 net/bluetooth/mgmt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/bluetooth/mgmt.c +2 -2
index 7e56cb3d377a1..b8e05ddeedba9 100644
@@ -5478,8 +5478,8 @@ static int remove_adv_monitor(struct sock *sk, struct hci_dev *hdev,
 		goto unlock;
 	}
 
-	err = hci_cmd_sync_queue(hdev, mgmt_remove_adv_monitor_sync, cmd,
-				 mgmt_remove_adv_monitor_complete);
+	err = hci_cmd_sync_submit(hdev, mgmt_remove_adv_monitor_sync, cmd,
+				  mgmt_remove_adv_monitor_complete);
 
 	if (err) {
 		mgmt_pending_remove(cmd);
Signature · 1 line
2.43.0
author maintainer
#91 lore raw

PATCH 91/228 Bluetooth: qca: set power_ctrl_enabled on NULL returned by gpiod_get_optional()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bartosz Golaszewski <[email protected]>

[ Upstream commit 3d05fc82237aa97162d0d7dc300b55bb34e91d02 ]

Any return value from gpiod_get_optional() other than a pointer to a GPIO descriptor or a NULL-pointer is an error and the driver should abort probing. That being said: commit 56d074d26c58 ("Bluetooth: hci_qca:
don't use IS_ERR_OR_NULL() with gpiod_get_optional()") no longer sets power_ctrl_enabled on NULL-pointer returned by devm_gpiod_get_optional(). Restore this behavior but bail-out on errors. While at it: also bail-out on error returned when trying to get the "swctrl" GPIO.

---
 drivers/bluetooth/hci_qca.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)
drivers/bluetooth/hci_qca.c +13 -5
index ecbc52eaf1010..2f7ae38d85eb0 100644
@@ -2329,16 +2329,21 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		    (data->soc_type == QCA_WCN6750 ||
 		     data->soc_type == QCA_WCN6855)) {
 			dev_err(&serdev->dev, "failed to acquire BT_EN gpio\n");
-			power_ctrl_enabled = false;
+			return PTR_ERR(qcadev->bt_en);
 		}
 
+		if (!qcadev->bt_en)
+			power_ctrl_enabled = false;
+
 		qcadev->sw_ctrl = devm_gpiod_get_optional(&serdev->dev, "swctrl",
 					       GPIOD_IN);
 		if (IS_ERR(qcadev->sw_ctrl) &&
 		    (data->soc_type == QCA_WCN6750 ||
 		     data->soc_type == QCA_WCN6855 ||
-		     data->soc_type == QCA_WCN7850))
-			dev_warn(&serdev->dev, "failed to acquire SW_CTRL gpio\n");
+		     data->soc_type == QCA_WCN7850)) {
+			dev_err(&serdev->dev, "failed to acquire SW_CTRL gpio\n");
+			return PTR_ERR(qcadev->sw_ctrl);
+		}
 
 		qcadev->susclk = devm_clk_get_optional(&serdev->dev, NULL);
 		if (IS_ERR(qcadev->susclk)) {
@@ -2357,10 +2362,13 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		qcadev->bt_en = devm_gpiod_get_optional(&serdev->dev, "enable",
 					       GPIOD_OUT_LOW);
 		if (IS_ERR(qcadev->bt_en)) {
-			dev_warn(&serdev->dev, "failed to acquire enable gpio\n");
-			power_ctrl_enabled = false;
+			dev_err(&serdev->dev, "failed to acquire enable gpio\n");
+			return PTR_ERR(qcadev->bt_en);
 		}
 
+		if (!qcadev->bt_en)
+			power_ctrl_enabled = false;
+
 		qcadev->susclk = devm_clk_get_optional(&serdev->dev, NULL);
 		if (IS_ERR(qcadev->susclk)) {
 			dev_warn(&serdev->dev, "failed to acquire clk\n");
Signature · 1 line
2.43.0
author maintainer
#92 lore raw

PATCH 92/228 ipvs: Fix checksumming on GSO of SCTP packets

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ismael Luceno <[email protected]>

[ Upstream commit e10d3ba4d434ed172914617ed8d74bd411421193 ]

It was observed in the wild that pairs of consecutive packets would leave the IPVS with the same wrong checksum, and the issue only went away when disabling GSO.

IPVS needs to avoid computing the SCTP checksum when using GSO.

---
 net/netfilter/ipvs/ip_vs_proto_sctp.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/netfilter/ipvs/ip_vs_proto_sctp.c +4 -2
index a0921adc31a9f..1e689c7141271 100644
@@ -126,7 +126,8 @@ sctp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 	if (sctph->source != cp->vport || payload_csum ||
 	    skb->ip_summed == CHECKSUM_PARTIAL) {
 		sctph->source = cp->vport;
-		sctp_nat_csum(skb, sctph, sctphoff);
+		if (!skb_is_gso(skb) || !skb_is_gso_sctp(skb))
+			sctp_nat_csum(skb, sctph, sctphoff);
 	} else {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
@@ -174,7 +175,8 @@ sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 	    (skb->ip_summed == CHECKSUM_PARTIAL &&
 	     !(skb_dst(skb)->dev->features & NETIF_F_SCTP_CRC))) {
 		sctph->dest = cp->dport;
-		sctp_nat_csum(skb, sctph, sctphoff);
+		if (!skb_is_gso(skb) || !skb_is_gso_sctp(skb))
+			sctp_nat_csum(skb, sctph, sctphoff);
 	} else if (skb->ip_summed != CHECKSUM_PARTIAL) {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
Signature · 1 line
2.43.0
author maintainer
#93 lore raw

PATCH 93/228 net: openvswitch: Fix Use-After-Free in ovs_ct_exit

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit 5ea7b72d4fac2fdbc0425cd8f2ea33abe95235b2 ]

Since kfree_rcu, which is called in the hlist_for_each_entry_rcu traversal of ovs_ct_limit_exit, is not part of the RCU read critical section, it is possible that the RCU grace period will pass during the traversal and the key will be free.

To prevent this, it should be changed to hlist_for_each_entry_safe.

---
 net/openvswitch/conntrack.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/openvswitch/conntrack.c +2 -2
index 74b63cdb59923..2928c142a2ddb 100644
@@ -1593,9 +1593,9 @@ static void ovs_ct_limit_exit(struct net *net, struct ovs_net *ovs_net)
 	for (i = 0; i < CT_LIMIT_HASH_BUCKETS; ++i) {
 		struct hlist_head *head = &info->limits[i];
 		struct ovs_ct_limit *ct_limit;
+		struct hlist_node *next;
 
-		hlist_for_each_entry_rcu(ct_limit, head, hlist_node,
-					 lockdep_ovsl_is_held())
+		hlist_for_each_entry_safe(ct_limit, next, head, hlist_node)
 			kfree_rcu(ct_limit, rcu);
 	}
 	kfree(info->limits);
Signature · 1 line
2.43.0
author maintainer
#94 lore raw

PATCH 94/228 mlxsw: Use refcount_t for reference counting

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Amit Cohen <[email protected]>

[ Upstream commit 1267f7223bec186dc26ef4e6075496c6217355de ]

mlxsw driver uses 'unsigned int' for reference counters in several structures. Instead, use refcount_t type which allows us to catch overflow and underflow issues. Change the type of the counters and use the appropriate API.

Stable-dep-of: 627f9c1bb882 ("mlxsw: spectrum_acl_tcam: Fix race in region ID allocation")

---
 .../mellanox/mlxsw/core_acl_flex_actions.c      | 16 ++++++++--------
 .../mellanox/mlxsw/core_acl_flex_keys.c         |  9 +++++----
 .../net/ethernet/mellanox/mlxsw/spectrum_acl.c  | 11 ++++++-----
 .../ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 17 +++++++++--------
 .../ethernet/mellanox/mlxsw/spectrum_router.c   | 15 ++++++++-------
 .../mellanox/mlxsw/spectrum_switchdev.c         |  8 ++++----
 6 files changed, 40 insertions(+), 36 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c +8 -8
index faa63ea9b83e1..1915fa41c6224 100644
@@ -95,7 +95,7 @@ struct mlxsw_afa_set {
 		      */
 	   has_trap:1,
 	   has_police:1;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	struct mlxsw_afa_set *next; /* Pointer to the next set. */
 	struct mlxsw_afa_set *prev; /* Pointer to the previous set,
 				     * note that set may have multiple
@@ -120,7 +120,7 @@ struct mlxsw_afa_fwd_entry {
 	struct rhash_head ht_node;
 	struct mlxsw_afa_fwd_entry_ht_key ht_key;
 	u32 kvdl_index;
-	unsigned int ref_count;
+	refcount_t ref_count;
 };
 
 static const struct rhashtable_params mlxsw_afa_fwd_entry_ht_params = {
@@ -282,7 +282,7 @@ static struct mlxsw_afa_set *mlxsw_afa_set_create(bool is_first)
 	/* Need to initialize the set to pass by default */
 	mlxsw_afa_set_goto_set(set, MLXSW_AFA_SET_GOTO_BINDING_CMD_TERM, 0);
 	set->ht_key.is_first = is_first;
-	set->ref_count = 1;
+	refcount_set(&set->ref_count, 1);
 	return set;
 }
 
@@ -330,7 +330,7 @@ static void mlxsw_afa_set_unshare(struct mlxsw_afa *mlxsw_afa,
 static void mlxsw_afa_set_put(struct mlxsw_afa *mlxsw_afa,
 			      struct mlxsw_afa_set *set)
 {
-	if (--set->ref_count)
+	if (!refcount_dec_and_test(&set->ref_count))
 		return;
 	if (set->shared)
 		mlxsw_afa_set_unshare(mlxsw_afa, set);
@@ -350,7 +350,7 @@ static struct mlxsw_afa_set *mlxsw_afa_set_get(struct mlxsw_afa *mlxsw_afa,
 	set = rhashtable_lookup_fast(&mlxsw_afa->set_ht, &orig_set->ht_key,
 				     mlxsw_afa_set_ht_params);
 	if (set) {
-		set->ref_count++;
+		refcount_inc(&set->ref_count);
 		mlxsw_afa_set_put(mlxsw_afa, orig_set);
 	} else {
 		set = orig_set;
@@ -564,7 +564,7 @@ mlxsw_afa_fwd_entry_create(struct mlxsw_afa *mlxsw_afa, u16 local_port)
 	if (!fwd_entry)
 		return ERR_PTR(-ENOMEM);
 	fwd_entry->ht_key.local_port = local_port;
-	fwd_entry->ref_count = 1;
+	refcount_set(&fwd_entry->ref_count, 1);
 
 	err = rhashtable_insert_fast(&mlxsw_afa->fwd_entry_ht,
 				     &fwd_entry->ht_node,
@@ -607,7 +607,7 @@ mlxsw_afa_fwd_entry_get(struct mlxsw_afa *mlxsw_afa, u16 local_port)
 	fwd_entry = rhashtable_lookup_fast(&mlxsw_afa->fwd_entry_ht, &ht_key,
 					   mlxsw_afa_fwd_entry_ht_params);
 	if (fwd_entry) {
-		fwd_entry->ref_count++;
+		refcount_inc(&fwd_entry->ref_count);
 		return fwd_entry;
 	}
 	return mlxsw_afa_fwd_entry_create(mlxsw_afa, local_port);
@@ -616,7 +616,7 @@ mlxsw_afa_fwd_entry_get(struct mlxsw_afa *mlxsw_afa, u16 local_port)
 static void mlxsw_afa_fwd_entry_put(struct mlxsw_afa *mlxsw_afa,
 				    struct mlxsw_afa_fwd_entry *fwd_entry)
 {
-	if (--fwd_entry->ref_count)
+	if (!refcount_dec_and_test(&fwd_entry->ref_count))
 		return;
 	mlxsw_afa_fwd_entry_destroy(mlxsw_afa, fwd_entry);
 }
drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c +5 -4
index 0d5e6f9b466ec..947500f8ed714 100644
@@ -5,6 +5,7 @@
 #include <linux/slab.h>
 #include <linux/list.h>
 #include <linux/errno.h>
+#include <linux/refcount.h>
 
 #include "item.h"
 #include "core_acl_flex_keys.h"
@@ -107,7 +108,7 @@ EXPORT_SYMBOL(mlxsw_afk_destroy);
 
 struct mlxsw_afk_key_info {
 	struct list_head list;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	unsigned int blocks_count;
 	int element_to_block[MLXSW_AFK_ELEMENT_MAX]; /* index is element, value
 						      * is index inside "blocks"
@@ -334,7 +335,7 @@ mlxsw_afk_key_info_create(struct mlxsw_afk *mlxsw_afk,
 	if (err)
 		goto err_picker;
 	list_add(&key_info->list, &mlxsw_afk->key_info_list);
-	key_info->ref_count = 1;
+	refcount_set(&key_info->ref_count, 1);
 	return key_info;
 
 err_picker:
@@ -356,7 +357,7 @@ mlxsw_afk_key_info_get(struct mlxsw_afk *mlxsw_afk,
 
 	key_info = mlxsw_afk_key_info_find(mlxsw_afk, elusage);
 	if (key_info) {
-		key_info->ref_count++;
+		refcount_inc(&key_info->ref_count);
 		return key_info;
 	}
 	return mlxsw_afk_key_info_create(mlxsw_afk, elusage);
@@ -365,7 +366,7 @@ EXPORT_SYMBOL(mlxsw_afk_key_info_get);
 
 void mlxsw_afk_key_info_put(struct mlxsw_afk_key_info *key_info)
 {
-	if (--key_info->ref_count)
+	if (!refcount_dec_and_test(&key_info->ref_count))
 		return;
 	mlxsw_afk_key_info_destroy(key_info);
 }
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c +6 -5
index 7c59c8a135840..b01b000bc71c1 100644
@@ -9,6 +9,7 @@
 #include <linux/rhashtable.h>
 #include <linux/netdevice.h>
 #include <linux/mutex.h>
+#include <linux/refcount.h>
 #include <net/net_namespace.h>
 #include <net/tc_act/tc_vlan.h>
 
@@ -55,7 +56,7 @@ struct mlxsw_sp_acl_ruleset {
 	struct rhash_head ht_node; /* Member of acl HT */
 	struct mlxsw_sp_acl_ruleset_ht_key ht_key;
 	struct rhashtable rule_ht;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	unsigned int min_prio;
 	unsigned int max_prio;
 	unsigned long priv[];
@@ -99,7 +100,7 @@ static bool
 mlxsw_sp_acl_ruleset_is_singular(const struct mlxsw_sp_acl_ruleset *ruleset)
 {
 	/* We hold a reference on ruleset ourselves */
-	return ruleset->ref_count == 2;
+	return refcount_read(&ruleset->ref_count) == 2;
 }
 
 int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
@@ -176,7 +177,7 @@ mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp,
 	ruleset = kzalloc(alloc_size, GFP_KERNEL);
 	if (!ruleset)
 		return ERR_PTR(-ENOMEM);
-	ruleset->ref_count = 1;
+	refcount_set(&ruleset->ref_count, 1);
 	ruleset->ht_key.block = block;
 	ruleset->ht_key.chain_index = chain_index;
 	ruleset->ht_key.ops = ops;
@@ -222,13 +223,13 @@ static void mlxsw_sp_acl_ruleset_destroy(struct mlxsw_sp *mlxsw_sp,
 
 static void mlxsw_sp_acl_ruleset_ref_inc(struct mlxsw_sp_acl_ruleset *ruleset)
 {
-	ruleset->ref_count++;
+	refcount_inc(&ruleset->ref_count);
 }
 
 static void mlxsw_sp_acl_ruleset_ref_dec(struct mlxsw_sp *mlxsw_sp,
 					 struct mlxsw_sp_acl_ruleset *ruleset)
 {
-	if (--ruleset->ref_count)
+	if (!refcount_dec_and_test(&ruleset->ref_count))
 		return;
 	mlxsw_sp_acl_ruleset_destroy(mlxsw_sp, ruleset);
 }
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +9 -8
index 50ea1eff02b2f..f20052776b3f2 100644
@@ -9,6 +9,7 @@
 #include <linux/rhashtable.h>
 #include <linux/netdevice.h>
 #include <linux/mutex.h>
+#include <linux/refcount.h>
 #include <net/devlink.h>
 #include <trace/events/mlxsw.h>
 
@@ -155,7 +156,7 @@ struct mlxsw_sp_acl_tcam_vregion {
 		struct mlxsw_sp_acl_tcam_rehash_ctx ctx;
 	} rehash;
 	struct mlxsw_sp *mlxsw_sp;
-	unsigned int ref_count;
+	refcount_t ref_count;
 };
 
 struct mlxsw_sp_acl_tcam_vchunk;
@@ -176,7 +177,7 @@ struct mlxsw_sp_acl_tcam_vchunk {
 	unsigned int priority; /* Priority within the vregion and group */
 	struct mlxsw_sp_acl_tcam_vgroup *vgroup;
 	struct mlxsw_sp_acl_tcam_vregion *vregion;
-	unsigned int ref_count;
+	refcount_t ref_count;
 };
 
 struct mlxsw_sp_acl_tcam_entry {
@@ -769,7 +770,7 @@ mlxsw_sp_acl_tcam_vregion_create(struct mlxsw_sp *mlxsw_sp,
 	vregion->tcam = tcam;
 	vregion->mlxsw_sp = mlxsw_sp;
 	vregion->vgroup = vgroup;
-	vregion->ref_count = 1;
+	refcount_set(&vregion->ref_count, 1);
 
 	vregion->key_info = mlxsw_afk_key_info_get(afk, elusage);
 	if (IS_ERR(vregion->key_info)) {
@@ -856,7 +857,7 @@ mlxsw_sp_acl_tcam_vregion_get(struct mlxsw_sp *mlxsw_sp,
 			 */
 			return ERR_PTR(-EOPNOTSUPP);
 		}
-		vregion->ref_count++;
+		refcount_inc(&vregion->ref_count);
 		return vregion;
 	}
 
@@ -871,7 +872,7 @@ static void
 mlxsw_sp_acl_tcam_vregion_put(struct mlxsw_sp *mlxsw_sp,
 			      struct mlxsw_sp_acl_tcam_vregion *vregion)
 {
-	if (--vregion->ref_count)
+	if (!refcount_dec_and_test(&vregion->ref_count))
 		return;
 	mlxsw_sp_acl_tcam_vregion_destroy(mlxsw_sp, vregion);
 }
@@ -924,7 +925,7 @@ mlxsw_sp_acl_tcam_vchunk_create(struct mlxsw_sp *mlxsw_sp,
 	INIT_LIST_HEAD(&vchunk->ventry_list);
 	vchunk->priority = priority;
 	vchunk->vgroup = vgroup;
-	vchunk->ref_count = 1;
+	refcount_set(&vchunk->ref_count, 1);
 
 	vregion = mlxsw_sp_acl_tcam_vregion_get(mlxsw_sp, vgroup,
 						priority, elusage);
@@ -1008,7 +1009,7 @@ mlxsw_sp_acl_tcam_vchunk_get(struct mlxsw_sp *mlxsw_sp,
 		if (WARN_ON(!mlxsw_afk_key_info_subset(vchunk->vregion->key_info,
 						       elusage)))
 			return ERR_PTR(-EINVAL);
-		vchunk->ref_count++;
+		refcount_inc(&vchunk->ref_count);
 		return vchunk;
 	}
 	return mlxsw_sp_acl_tcam_vchunk_create(mlxsw_sp, vgroup,
@@ -1019,7 +1020,7 @@ static void
 mlxsw_sp_acl_tcam_vchunk_put(struct mlxsw_sp *mlxsw_sp,
 			     struct mlxsw_sp_acl_tcam_vchunk *vchunk)
 {
-	if (--vchunk->ref_count)
+	if (!refcount_dec_and_test(&vchunk->ref_count))
 		return;
 	mlxsw_sp_acl_tcam_vchunk_destroy(mlxsw_sp, vchunk);
 }
drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +8 -7
index 7164f9e6370fb..87617df694ab2 100644
@@ -501,7 +501,7 @@ struct mlxsw_sp_rt6 {
 
 struct mlxsw_sp_lpm_tree {
 	u8 id; /* tree ID */
-	unsigned int ref_count;
+	refcount_t ref_count;
 	enum mlxsw_sp_l3proto proto;
 	unsigned long prefix_ref_count[MLXSW_SP_PREFIX_COUNT];
 	struct mlxsw_sp_prefix_usage prefix_usage;
@@ -578,7 +578,7 @@ mlxsw_sp_lpm_tree_find_unused(struct mlxsw_sp *mlxsw_sp)
 
 	for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) {
 		lpm_tree = &mlxsw_sp->router->lpm.trees[i];
-		if (lpm_tree->ref_count == 0)
+		if (refcount_read(&lpm_tree->ref_count) == 0)
 			return lpm_tree;
 	}
 	return NULL;
@@ -654,7 +654,7 @@ mlxsw_sp_lpm_tree_create(struct mlxsw_sp *mlxsw_sp,
 	       sizeof(lpm_tree->prefix_usage));
 	memset(&lpm_tree->prefix_ref_count, 0,
 	       sizeof(lpm_tree->prefix_ref_count));
-	lpm_tree->ref_count = 1;
+	refcount_set(&lpm_tree->ref_count, 1);
 	return lpm_tree;
 
 err_left_struct_set:
@@ -678,7 +678,7 @@ mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp,
 
 	for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) {
 		lpm_tree = &mlxsw_sp->router->lpm.trees[i];
-		if (lpm_tree->ref_count != 0 &&
+		if (refcount_read(&lpm_tree->ref_count) &&
 		    lpm_tree->proto == proto &&
 		    mlxsw_sp_prefix_usage_eq(&lpm_tree->prefix_usage,
 					     prefix_usage)) {
@@ -691,14 +691,15 @@ mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp,
 
 static void mlxsw_sp_lpm_tree_hold(struct mlxsw_sp_lpm_tree *lpm_tree)
 {
-	lpm_tree->ref_count++;
+	refcount_inc(&lpm_tree->ref_count);
 }
 
 static void mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp,
 				  struct mlxsw_sp_lpm_tree *lpm_tree)
 {
-	if (--lpm_tree->ref_count == 0)
-		mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree);
+	if (!refcount_dec_and_test(&lpm_tree->ref_count))
+		return;
+	mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree);
 }
 
 #define MLXSW_SP_LPM_TREE_MIN 1 /* tree 0 is reserved */
drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +4 -4
index 6c749c148148d..6397ff0dc951c 100644
@@ -61,7 +61,7 @@ struct mlxsw_sp_bridge_port {
 	struct mlxsw_sp_bridge_device *bridge_device;
 	struct list_head list;
 	struct list_head vlans_list;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	u8 stp_state;
 	unsigned long flags;
 	bool mrouter;
@@ -495,7 +495,7 @@ mlxsw_sp_bridge_port_create(struct mlxsw_sp_bridge_device *bridge_device,
 			     BR_MCAST_FLOOD;
 	INIT_LIST_HEAD(&bridge_port->vlans_list);
 	list_add(&bridge_port->list, &bridge_device->ports_list);
-	bridge_port->ref_count = 1;
+	refcount_set(&bridge_port->ref_count, 1);
 
 	err = switchdev_bridge_port_offload(brport_dev, mlxsw_sp_port->dev,
 					    NULL, NULL, NULL, false, extack);
@@ -531,7 +531,7 @@ mlxsw_sp_bridge_port_get(struct mlxsw_sp_bridge *bridge,
 
 	bridge_port = mlxsw_sp_bridge_port_find(bridge, brport_dev);
 	if (bridge_port) {
-		bridge_port->ref_count++;
+		refcount_inc(&bridge_port->ref_count);
 		return bridge_port;
 	}
 
@@ -558,7 +558,7 @@ static void mlxsw_sp_bridge_port_put(struct mlxsw_sp_bridge *bridge,
 {
 	struct mlxsw_sp_bridge_device *bridge_device;
 
-	if (--bridge_port->ref_count != 0)
+	if (!refcount_dec_and_test(&bridge_port->ref_count))
 		return;
 	bridge_device = bridge_port->bridge_device;
 	mlxsw_sp_bridge_port_destroy(bridge_port);
Signature · 1 line
2.43.0
author maintainer
#95 lore raw

PATCH 95/228 mlxsw: spectrum_acl_tcam: Fix race in region ID allocation

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 627f9c1bb882765a84aa78015abbacd783d429be ]

Region identifiers can be allocated both when user space tries to insert a new tc filter and when filters are migrated from one region to another as part of the rehash delayed work.

There is no lock protecting the bitmap from which these identifiers are allocated from, which is racy and leads to bad parameter errors from the device's firmware.

Fix by converting the bitmap to IDA which handles its own locking. For consistency, do the same for the group identifiers that are part of the same structure.

---
 .../mellanox/mlxsw/spectrum_acl_tcam.c        | 61 ++++++++-----------
 .../mellanox/mlxsw/spectrum_acl_tcam.h        |  5 +-
 2 files changed, 30 insertions(+), 36 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +27 -34
index f20052776b3f2..b6a4652a6475a 100644
@@ -10,6 +10,7 @@
 #include <linux/netdevice.h>
 #include <linux/mutex.h>
 #include <linux/refcount.h>
+#include <linux/idr.h>
 #include <net/devlink.h>
 #include <trace/events/mlxsw.h>
 
@@ -58,41 +59,43 @@ int mlxsw_sp_acl_tcam_priority_get(struct mlxsw_sp *mlxsw_sp,
 static int mlxsw_sp_acl_tcam_region_id_get(struct mlxsw_sp_acl_tcam *tcam,
 					   u16 *p_id)
 {
-	u16 id;
+	int id;
 
-	id = find_first_zero_bit(tcam->used_regions, tcam->max_regions);
-	if (id < tcam->max_regions) {
-		__set_bit(id, tcam->used_regions);
-		*p_id = id;
-		return 0;
-	}
-	return -ENOBUFS;
+	id = ida_alloc_max(&tcam->used_regions, tcam->max_regions - 1,
+			   GFP_KERNEL);
+	if (id < 0)
+		return id;
+
+	*p_id = id;
+
+	return 0;
 }
 
 static void mlxsw_sp_acl_tcam_region_id_put(struct mlxsw_sp_acl_tcam *tcam,
 					    u16 id)
 {
-	__clear_bit(id, tcam->used_regions);
+	ida_free(&tcam->used_regions, id);
 }
 
 static int mlxsw_sp_acl_tcam_group_id_get(struct mlxsw_sp_acl_tcam *tcam,
 					  u16 *p_id)
 {
-	u16 id;
+	int id;
 
-	id = find_first_zero_bit(tcam->used_groups, tcam->max_groups);
-	if (id < tcam->max_groups) {
-		__set_bit(id, tcam->used_groups);
-		*p_id = id;
-		return 0;
-	}
-	return -ENOBUFS;
+	id = ida_alloc_max(&tcam->used_groups, tcam->max_groups - 1,
+			   GFP_KERNEL);
+	if (id < 0)
+		return id;
+
+	*p_id = id;
+
+	return 0;
 }
 
 static void mlxsw_sp_acl_tcam_group_id_put(struct mlxsw_sp_acl_tcam *tcam,
 					   u16 id)
 {
-	__clear_bit(id, tcam->used_groups);
+	ida_free(&tcam->used_groups, id);
 }
 
 struct mlxsw_sp_acl_tcam_pattern {
@@ -1549,19 +1552,11 @@ int mlxsw_sp_acl_tcam_init(struct mlxsw_sp *mlxsw_sp,
 	if (max_tcam_regions < max_regions)
 		max_regions = max_tcam_regions;
 
-	tcam->used_regions = bitmap_zalloc(max_regions, GFP_KERNEL);
-	if (!tcam->used_regions) {
-		err = -ENOMEM;
-		goto err_alloc_used_regions;
-	}
+	ida_init(&tcam->used_regions);
 	tcam->max_regions = max_regions;
 
 	max_groups = MLXSW_CORE_RES_GET(mlxsw_sp->core, ACL_MAX_GROUPS);
-	tcam->used_groups = bitmap_zalloc(max_groups, GFP_KERNEL);
-	if (!tcam->used_groups) {
-		err = -ENOMEM;
-		goto err_alloc_used_groups;
-	}
+	ida_init(&tcam->used_groups);
 	tcam->max_groups = max_groups;
 	tcam->max_group_size = MLXSW_CORE_RES_GET(mlxsw_sp->core,
 						  ACL_MAX_GROUP_SIZE);
@@ -1575,10 +1570,8 @@ int mlxsw_sp_acl_tcam_init(struct mlxsw_sp *mlxsw_sp,
 	return 0;
 
 err_tcam_init:
-	bitmap_free(tcam->used_groups);
-err_alloc_used_groups:
-	bitmap_free(tcam->used_regions);
-err_alloc_used_regions:
+	ida_destroy(&tcam->used_groups);
+	ida_destroy(&tcam->used_regions);
 	mlxsw_sp_acl_tcam_rehash_params_unregister(mlxsw_sp);
 err_rehash_params_register:
 	mutex_destroy(&tcam->lock);
@@ -1591,8 +1584,8 @@ void mlxsw_sp_acl_tcam_fini(struct mlxsw_sp *mlxsw_sp,
 	const struct mlxsw_sp_acl_tcam_ops *ops = mlxsw_sp->acl_tcam_ops;
 
 	ops->fini(mlxsw_sp, tcam->priv);
-	bitmap_free(tcam->used_groups);
-	bitmap_free(tcam->used_regions);
+	ida_destroy(&tcam->used_groups);
+	ida_destroy(&tcam->used_regions);
 	mlxsw_sp_acl_tcam_rehash_params_unregister(mlxsw_sp);
 	mutex_destroy(&tcam->lock);
 }
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h +3 -2
index 462bf448497d3..79a1d86065125 100644
@@ -6,15 +6,16 @@
 
 #include <linux/list.h>
 #include <linux/parman.h>
+#include <linux/idr.h>
 
 #include "reg.h"
 #include "spectrum.h"
 #include "core_acl_flex_keys.h"
 
 struct mlxsw_sp_acl_tcam {
-	unsigned long *used_regions; /* bit array */
+	struct ida used_regions;
 	unsigned int max_regions;
-	unsigned long *used_groups;  /* bit array */
+	struct ida used_groups;
 	unsigned int max_groups;
 	unsigned int max_group_size;
 	struct mutex lock; /* guards vregion list */
Signature · 1 line
2.43.0
author maintainer
#96 lore raw

PATCH 96/228 mlxsw: spectrum_acl_tcam: Fix race during rehash delayed work

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit d90cfe20562407d9f080d24123078d666d730707 ]

The purpose of the rehash delayed work is to reduce the number of masks (eRPs) used by an ACL region as the eRP bank is a global and limited resource.

This is done in three steps:

1. Creating a new set of masks and a new ACL region which will use the new masks and to which the existing filters will be migrated to. The new region is assigned to 'vregion->region' and the region from which the filters are migrated from is assigned to 'vregion->region2'.

2. Migrating all the filters from the old region to the new region.

3. Destroying the old region and setting 'vregion->region2' to NULL.

Only the second steps is performed under the 'vregion->lock' mutex although its comments says that among other things it "Protects consistency of region, region2 pointers".

This is problematic as the first step can race with filter insertion from user space that uses 'vregion->region', but under the mutex.

Fix by holding the mutex across the entirety of the delayed work and not only during the second step.

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +2 -2
index b6a4652a6475a..9c0c728bb42dc 100644
@@ -718,7 +718,9 @@ static void mlxsw_sp_acl_tcam_vregion_rehash_work(struct work_struct *work)
 			     rehash.dw.work);
 	int credits = MLXSW_SP_ACL_TCAM_VREGION_REHASH_CREDITS;
 
+	mutex_lock(&vregion->lock);
 	mlxsw_sp_acl_tcam_vregion_rehash(vregion->mlxsw_sp, vregion, &credits);
+	mutex_unlock(&vregion->lock);
 	if (credits < 0)
 		/* Rehash gone out of credits so it was interrupted.
 		 * Schedule the work as soon as possible to continue.
@@ -1323,7 +1325,6 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 	int err, err2;
 
 	trace_mlxsw_sp_acl_tcam_vregion_migrate(mlxsw_sp, vregion);
-	mutex_lock(&vregion->lock);
 	err = mlxsw_sp_acl_tcam_vchunk_migrate_all(mlxsw_sp, vregion,
 						   ctx, credits);
 	if (err) {
@@ -1343,7 +1344,6 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 			/* Let the rollback to be continued later on. */
 		}
 	}
-	mutex_unlock(&vregion->lock);
 	trace_mlxsw_sp_acl_tcam_vregion_migrate_end(mlxsw_sp, vregion);
 	return err;
 }
Signature · 1 line
2.43.0
author maintainer
#97 lore raw

PATCH 97/228 mlxsw: spectrum_acl_tcam: Fix possible use-after-free during activity update

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 79b5b4b18bc85b19d3a518483f9abbbe6d7b3ba4 ]

The rule activity update delayed work periodically traverses the list of configured rules and queries their activity from the device.

As part of this task it accesses the entry pointed by 'ventry->entry', but this entry can be changed concurrently by the rehash delayed work, leading to a use-after-free [1].

Fix by closing the race and perform the activity query under the 'vregion->lock' mutex.

[1]

BUG: KASAN: slab-use-after-free in mlxsw_sp_acl_tcam_flower_rule_activity_get+0x121/0x140
Read of size 8 at addr ffff8881054ed808 by task kworker/0:18/181

CPU: 0 PID: 181 Comm: kworker/0:18 Not tainted 6.9.0-rc2-custom-00781-gd5ab772d32f7 #2
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_rule_activity_update_work
Call Trace:
 <TASK>
 dump_stack_lvl+0xc6/0x120
 print_report+0xce/0x670
 kasan_report+0xd7/0x110
 mlxsw_sp_acl_tcam_flower_rule_activity_get+0x121/0x140
 mlxsw_sp_acl_rule_activity_update_work+0x219/0x400
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30
 </TASK>

Allocated by task 1039:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 __kasan_kmalloc+0x8f/0xa0
 __kmalloc+0x19c/0x360
 mlxsw_sp_acl_tcam_entry_create+0x7b/0x1f0
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x30d/0xb50
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x157/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30

Freed by task 1039:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 kasan_save_free_info+0x3b/0x60
 poison_slab_object+0x102/0x170
 __kasan_slab_free+0x14/0x30
 kfree+0xc1/0x290
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x3d7/0xb50
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x157/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30
---
 .../net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c    | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +8 -2
index 9c0c728bb42dc..7e69225c057de 100644
@@ -1159,8 +1159,14 @@ mlxsw_sp_acl_tcam_ventry_activity_get(struct mlxsw_sp *mlxsw_sp,
 				      struct mlxsw_sp_acl_tcam_ventry *ventry,
 				      bool *activity)
 {
-	return mlxsw_sp_acl_tcam_entry_activity_get(mlxsw_sp,
-						    ventry->entry, activity);
+	struct mlxsw_sp_acl_tcam_vregion *vregion = ventry->vchunk->vregion;
+	int err;
+
+	mutex_lock(&vregion->lock);
+	err = mlxsw_sp_acl_tcam_entry_activity_get(mlxsw_sp, ventry->entry,
+						   activity);
+	mutex_unlock(&vregion->lock);
+	return err;
 }
 
 static int
Signature · 1 line
2.43.0
author maintainer
#98 lore raw

PATCH 98/228 mlxsw: spectrum_acl_tcam: Fix possible use-after-free during rehash

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 54225988889931467a9b55fdbef534079b665519 ]

The rehash delayed work migrates filters from one region to another according to the number of available credits.

The migrated from region is destroyed at the end of the work if the number of credits is non-negative as the assumption is that this is indicative of migration being complete. This assumption is incorrect as a non-negative number of credits can also be the result of a failed migration.

The destruction of a region that still has filters referencing it can result in a use-after-free [1].

Fix by not destroying the region if migration failed.

[1]

BUG: KASAN: slab-use-after-free in mlxsw_sp_acl_ctcam_region_entry_remove+0x21d/0x230
Read of size 8 at addr ffff8881735319e8 by task kworker/0:31/3858

CPU: 0 PID: 3858 Comm: kworker/0:31 Tainted: G        W          6.9.0-rc2-custom-00782-gf2275c2157d8 #5
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
Call Trace:
 <TASK>
 dump_stack_lvl+0xc6/0x120
 print_report+0xce/0x670
 kasan_report+0xd7/0x110
 mlxsw_sp_acl_ctcam_region_entry_remove+0x21d/0x230
 mlxsw_sp_acl_ctcam_entry_del+0x2e/0x70
 mlxsw_sp_acl_atcam_entry_del+0x81/0x210
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x3cd/0xb50
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x157/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30
 </TASK>

Allocated by task 174:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 __kasan_kmalloc+0x8f/0xa0
 __kmalloc+0x19c/0x360
 mlxsw_sp_acl_tcam_region_create+0xdf/0x9c0
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x954/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30

Freed by task 7:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 kasan_save_free_info+0x3b/0x60
 poison_slab_object+0x102/0x170
 __kasan_slab_free+0x14/0x30
 kfree+0xc1/0x290
 mlxsw_sp_acl_tcam_region_destroy+0x272/0x310
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x731/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +1 -0
index 7e69225c057de..1ff0b2c7c11de 100644
@@ -1451,6 +1451,7 @@ mlxsw_sp_acl_tcam_vregion_rehash(struct mlxsw_sp *mlxsw_sp,
 						ctx, credits);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Failed to migrate vregion\n");
+		return;
 	}
 
 	if (*credits >= 0)
Signature · 1 line
2.43.0
author maintainer
#99 lore raw

PATCH 99/228 mlxsw: spectrum_acl_tcam: Rate limit error message

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 5bcf925587e9b5d36420d572a0b4d131c90fb306 ]

In the rare cases when the device resources are exhausted it is likely that the rehash delayed work will fail. An error message will be printed whenever this happens which can be overwhelming considering the fact that the work is per-region and that there can be hundreds of regions.

Fix by rate limiting the error message.

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +1 -1
index 1ff0b2c7c11de..568ae7092fe0e 100644
@@ -1450,7 +1450,7 @@ mlxsw_sp_acl_tcam_vregion_rehash(struct mlxsw_sp *mlxsw_sp,
 	err = mlxsw_sp_acl_tcam_vregion_migrate(mlxsw_sp, vregion,
 						ctx, credits);
 	if (err) {
-		dev_err(mlxsw_sp->bus_info->dev, "Failed to migrate vregion\n");
+		dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to migrate vregion\n");
 		return;
 	}
 
Signature · 1 line
2.43.0
author maintainer
#100 lore raw

PATCH 100/228 mlxsw: spectrum_acl_tcam: Fix memory leak during rehash

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 8ca3f7a7b61393804c46f170743c3b839df13977 ]

The rehash delayed work migrates filters from one region to another. This is done by iterating over all chunks (all the filters with the same priority) in the region and in each chunk iterating over all the filters.

If the migration fails, the code tries to migrate the filters back to the old region. However, the rollback itself can also fail in which case another migration will be erroneously performed. Besides the fact that this ping pong is not a very good idea, it also creates a problem.

Each virtual chunk references two chunks: The currently used one ('vchunk->chunk') and a backup ('vchunk->chunk2'). During migration the first holds the chunk we want to migrate filters to and the second holds the chunk we are migrating filters from.

The code currently assumes - but does not verify - that the backup chunk does not exist (NULL) if the currently used chunk does not reference the target region. This assumption breaks when we are trying to rollback a rollback, resulting in the backup chunk being overwritten and leaked [1].

Fix by not rolling back a failed rollback and add a warning to avoid future cases.

[1]

WARNING: CPU: 5 PID: 1063 at lib/parman.c:291 parman_destroy+0x17/0x20
Modules linked in:
CPU: 5 PID: 1063 Comm: kworker/5:11 Tainted: G        W          6.9.0-rc2-custom-00784-gc6a05c468a0b #14
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:parman_destroy+0x17/0x20
[...]
Call Trace:
 <TASK>
 mlxsw_sp_acl_atcam_region_fini+0x19/0x60
 mlxsw_sp_acl_tcam_region_destroy+0x49/0xf0
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x1f1/0x470
 process_one_work+0x151/0x370
 worker_thread+0x2cb/0x3e0
 kthread+0xd0/0x100
 ret_from_fork+0x34/0x50
 ret_from_fork_asm+0x1a/0x30
 </TASK>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +4 -0
index 568ae7092fe0e..0902eb7651e14 100644
@@ -1200,6 +1200,8 @@ mlxsw_sp_acl_tcam_vchunk_migrate_start(struct mlxsw_sp *mlxsw_sp,
 {
 	struct mlxsw_sp_acl_tcam_chunk *new_chunk;
 
+	WARN_ON(vchunk->chunk2);
+
 	new_chunk = mlxsw_sp_acl_tcam_chunk_create(mlxsw_sp, vchunk, region);
 	if (IS_ERR(new_chunk))
 		return PTR_ERR(new_chunk);
@@ -1334,6 +1336,8 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 	err = mlxsw_sp_acl_tcam_vchunk_migrate_all(mlxsw_sp, vregion,
 						   ctx, credits);
 	if (err) {
+		if (ctx->this_is_rollback)
+			return err;
 		/* In case migration was not successful, we need to swap
 		 * so the original region pointer is assigned again
 		 * to vregion->region.
Signature · 1 line
2.43.0
author maintainer
#101 lore raw

PATCH 101/228 mlxsw: spectrum_acl_tcam: Fix warning during rehash

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 743edc8547a92b6192aa1f1b6bb78233fa21dc9b ]

As previously explained, the rehash delayed work migrates filters from one region to another. This is done by iterating over all chunks (all the filters with the same priority) in the region and in each chunk iterating over all the filters.

When the work runs out of credits it stores the current chunk and entry as markers in the per-work context so that it would know where to resume the migration from the next time the work is scheduled.

Upon error, the chunk marker is reset to NULL, but without resetting the entry markers despite being relative to it. This can result in migration being resumed from an entry that does not belong to the chunk being migrated. In turn, this will eventually lead to a chunk being iterated over as if it is an entry. Because of how the two structures happen to be defined, this does not lead to KASAN splats, but to warnings such as [1].

Fix by creating a helper that resets all the markers and call it from all the places the currently only reset the chunk marker. For good measures also call it when starting a completely new rehash. Add a warning to avoid future cases.

[1]

WARNING: CPU: 7 PID: 1076 at drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c:407 mlxsw_afk_encode+0x242/0x2f0
Modules linked in:
CPU: 7 PID: 1076 Comm: kworker/7:24 Tainted: G        W          6.9.0-rc3-custom-00880-g29e61d91b77b #29
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:mlxsw_afk_encode+0x242/0x2f0
[...]
Call Trace:
 <TASK>
 mlxsw_sp_acl_atcam_entry_add+0xd9/0x3c0
 mlxsw_sp_acl_tcam_entry_create+0x5e/0xa0
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x109/0x290
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x6c/0x470
 process_one_work+0x151/0x370
 worker_thread+0x2cb/0x3e0
 kthread+0xd0/0x100
 ret_from_fork+0x34/0x50
 </TASK>
---
 .../mellanox/mlxsw/spectrum_acl_tcam.c        | 20 ++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +17 -3
index 0902eb7651e14..e8c6078866213 100644
@@ -730,6 +730,17 @@ static void mlxsw_sp_acl_tcam_vregion_rehash_work(struct work_struct *work)
 		mlxsw_sp_acl_tcam_vregion_rehash_work_schedule(vregion);
 }
 
+static void
+mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(struct mlxsw_sp_acl_tcam_rehash_ctx *ctx)
+{
+	/* The entry markers are relative to the current chunk and therefore
+	 * needs to be reset together with the chunk marker.
+	 */
+	ctx->current_vchunk = NULL;
+	ctx->start_ventry = NULL;
+	ctx->stop_ventry = NULL;
+}
+
 static void
 mlxsw_sp_acl_tcam_rehash_ctx_vchunk_changed(struct mlxsw_sp_acl_tcam_vchunk *vchunk)
 {
@@ -752,7 +763,7 @@ mlxsw_sp_acl_tcam_rehash_ctx_vregion_changed(struct mlxsw_sp_acl_tcam_vregion *v
 	 * the current chunk pointer to make sure all chunks
 	 * are properly migrated.
 	 */
-	vregion->rehash.ctx.current_vchunk = NULL;
+	mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(&vregion->rehash.ctx);
 }
 
 static struct mlxsw_sp_acl_tcam_vregion *
@@ -1220,7 +1231,7 @@ mlxsw_sp_acl_tcam_vchunk_migrate_end(struct mlxsw_sp *mlxsw_sp,
 {
 	mlxsw_sp_acl_tcam_chunk_destroy(mlxsw_sp, vchunk->chunk2);
 	vchunk->chunk2 = NULL;
-	ctx->current_vchunk = NULL;
+	mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(ctx);
 }
 
 static int
@@ -1252,6 +1263,8 @@ mlxsw_sp_acl_tcam_vchunk_migrate_one(struct mlxsw_sp *mlxsw_sp,
 		ventry = list_first_entry(&vchunk->ventry_list,
 					  typeof(*ventry), list);
 
+	WARN_ON(ventry->vchunk != vchunk);
+
 	list_for_each_entry_from(ventry, &vchunk->ventry_list, list) {
 		/* During rollback, once we reach the ventry that failed
 		 * to migrate, we are done.
@@ -1343,7 +1356,7 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 		 * to vregion->region.
 		 */
 		swap(vregion->region, vregion->region2);
-		ctx->current_vchunk = NULL;
+		mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(ctx);
 		ctx->this_is_rollback = true;
 		err2 = mlxsw_sp_acl_tcam_vchunk_migrate_all(mlxsw_sp, vregion,
 							    ctx, credits);
@@ -1402,6 +1415,7 @@ mlxsw_sp_acl_tcam_vregion_rehash_start(struct mlxsw_sp *mlxsw_sp,
 
 	ctx->hints_priv = hints_priv;
 	ctx->this_is_rollback = false;
+	mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(ctx);
 
 	return 0;
 
Signature · 1 line
2.43.0
author maintainer
#102 lore raw

PATCH 102/228 mlxsw: spectrum_acl_tcam: Fix incorrect list API usage

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit b377add0f0117409c418ddd6504bd682ebe0bf79 ]

Both the function that migrates all the chunks within a region and the function that migrates all the entries within a chunk call list_first_entry() on the respective lists without checking that the lists are not empty. This is incorrect usage of the API, which leads to the following warning [1].

Fix by returning if the lists are empty as there is nothing to migrate in this case.

[1]

WARNING: CPU: 0 PID: 6437 at drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c:1266 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x1f1/0>
Modules linked in:
CPU: 0 PID: 6437 Comm: kworker/0:37 Not tainted 6.9.0-rc3-custom-00883-g94a65f079ef6 #39
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:mlxsw_sp_acl_tcam_vchunk_migrate_all+0x1f1/0x2c0
[...]
Call Trace:
 <TASK>
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x6c/0x4a0
 process_one_work+0x151/0x370
 worker_thread+0x2cb/0x3e0
 kthread+0xd0/0x100
 ret_from_fork+0x34/0x50
 ret_from_fork_asm+0x1a/0x30
 </TASK>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +7 -0
index e8c6078866213..89a5ebc3463ff 100644
@@ -1254,6 +1254,9 @@ mlxsw_sp_acl_tcam_vchunk_migrate_one(struct mlxsw_sp *mlxsw_sp,
 		return 0;
 	}
 
+	if (list_empty(&vchunk->ventry_list))
+		goto out;
+
 	/* If the migration got interrupted, we have the ventry to start from
 	 * stored in context.
 	 */
@@ -1305,6 +1308,7 @@ mlxsw_sp_acl_tcam_vchunk_migrate_one(struct mlxsw_sp *mlxsw_sp,
 		}
 	}
 
+out:
 	mlxsw_sp_acl_tcam_vchunk_migrate_end(mlxsw_sp, vchunk, ctx);
 	return 0;
 }
@@ -1318,6 +1322,9 @@ mlxsw_sp_acl_tcam_vchunk_migrate_all(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_acl_tcam_vchunk *vchunk;
 	int err;
 
+	if (list_empty(&vregion->vchunk_list))
+		return 0;
+
 	/* If the migration got interrupted, we have the vchunk
 	 * we are working on stored in context.
 	 */
Signature · 1 line
2.43.0
author maintainer
#103 lore raw

PATCH 103/228 mlxsw: spectrum_acl_tcam: Fix memory leak when canceling rehash work

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit fb4e2b70a7194b209fc7320bbf33b375f7114bd5 ]

The rehash delayed work is rescheduled with a delay if the number of credits at end of the work is not negative as supposedly it means that the migration ended. Otherwise, it is rescheduled immediately.

After "mlxsw: spectrum_acl_tcam: Fix possible use-after-free during rehash" the above is no longer accurate as a non-negative number of credits is no longer indicative of the migration being done. It can also happen if the work encountered an error in which case the migration will resume the next time the work is scheduled.

The significance of the above is that it is possible for the work to be pending and associated with hints that were allocated when the migration started. This leads to the hints being leaked [1] when the work is canceled while pending as part of ACL region dismantle.

Fix by freeing the hints if hints are associated with a work that was canceled while pending.

Blame the original commit since the reliance on not having a pending work associated with hints is fragile.

[1]
unreferenced object 0xffff88810e7c3000 (size 256):
comm "kworker/0:16", pid 176, jiffies 4295460353 hex dump (first 32 bytes):

    00 30 95 11 81 88 ff ff 61 00 00 00 00 00 00 80  .0......a.......
    00 00 61 00 40 00 00 00 00 00 00 00 04 00 00 00  ..a.@...........
  backtrace (crc 2544ddb9):
    [<00000000cf8cfab3>] kmalloc_trace+0x23f/0x2a0
    [<000000004d9a1ad9>] objagg_hints_get+0x42/0x390
    [<000000000b143cf3>] mlxsw_sp_acl_erp_rehash_hints_get+0xca/0x400
    [<0000000059bdb60a>] mlxsw_sp_acl_tcam_vregion_rehash_work+0x868/0x1160
    [<00000000e81fd734>] process_one_work+0x59c/0xf20
    [<00000000ceee9e81>] worker_thread+0x799/0x12c0
    [<00000000bda6fe39>] kthread+0x246/0x300
    [<0000000070056d23>] ret_from_fork+0x34/0x70
    [<00000000dea2b93e>] ret_from_fork_asm+0x1a/0x30
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +5 -1
index 89a5ebc3463ff..92a406f02eae7 100644
@@ -836,10 +836,14 @@ mlxsw_sp_acl_tcam_vregion_destroy(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_acl_tcam *tcam = vregion->tcam;
 
 	if (vgroup->vregion_rehash_enabled && ops->region_rehash_hints_get) {
+		struct mlxsw_sp_acl_tcam_rehash_ctx *ctx = &vregion->rehash.ctx;
+
 		mutex_lock(&tcam->lock);
 		list_del(&vregion->tlist);
 		mutex_unlock(&tcam->lock);
-		cancel_delayed_work_sync(&vregion->rehash.dw);
+		if (cancel_delayed_work_sync(&vregion->rehash.dw) &&
+		    ctx->hints_priv)
+			ops->region_rehash_hints_put(ctx->hints_priv);
 	}
 	mlxsw_sp_acl_tcam_vgroup_vregion_detach(mlxsw_sp, vregion);
 	if (vregion->region2)
Signature · 1 line
2.43.0
author maintainer
#104 lore raw

PATCH 104/228 eth: bnxt: fix counting packets discarded due to OOM and netpoll

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jakub Kicinski <[email protected]>

[ Upstream commit 730117730709992c9f6535dd7b47638ee561ec45 ]

I added OOM and netpoll discard counters, naively assuming that the cpr pointer is pointing to a common completion ring. Turns out that is usually *a* completion ring but not *the* completion ring which bnapi->cp_ring points to. bnapi->cp_ring is where the stats are read from, so we end up reporting 0 thru ethtool -S and qstat even though the drop events have happened. Make 100% sure we're recording statistics in the correct structure.

Fixes 907fd4a294db ("bnxt: count discards due to memory allocation errors")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 44 ++++++++++-------------
 1 file changed, 18 insertions(+), 26 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +18 -26
index 5b4d810748f2a..e133a412a13fc 100644
@@ -1735,7 +1735,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		skb = bnxt_copy_skb(bnapi, data_ptr, len, mapping);
 		if (!skb) {
 			bnxt_abort_tpa(cpr, idx, agg_bufs);
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}
 	} else {
@@ -1745,7 +1745,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		new_data = __bnxt_alloc_rx_frag(bp, &new_mapping, GFP_ATOMIC);
 		if (!new_data) {
 			bnxt_abort_tpa(cpr, idx, agg_bufs);
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}
 
@@ -1761,7 +1761,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		if (!skb) {
 			skb_free_frag(data);
 			bnxt_abort_tpa(cpr, idx, agg_bufs);
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}
 		skb_reserve(skb, bp->rx_offset);
@@ -1772,7 +1772,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		skb = bnxt_rx_agg_pages_skb(bp, cpr, skb, idx, agg_bufs, true);
 		if (!skb) {
 			/* Page reuse already handled by bnxt_rx_pages(). */
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}
 	}
@@ -2051,11 +2051,8 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 			u32 frag_len = bnxt_rx_agg_pages_xdp(bp, cpr, &xdp,
 							     cp_cons, agg_bufs,
 							     false);
-			if (!frag_len) {
-				cpr->sw_stats.rx.rx_oom_discards += 1;
-				rc = -ENOMEM;
-				goto next_rx;
-			}
+			if (!frag_len)
+				goto oom_next_rx;
 		}
 		xdp_active = true;
 	}
@@ -2078,9 +2075,7 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 				else
 					bnxt_xdp_buff_frags_free(rxr, &xdp);
 			}
-			cpr->sw_stats.rx.rx_oom_discards += 1;
-			rc = -ENOMEM;
-			goto next_rx;
+			goto oom_next_rx;
 		}
 	} else {
 		u32 payload;
@@ -2091,29 +2086,21 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 			payload = 0;
 		skb = bp->rx_skb_func(bp, rxr, cons, data, data_ptr, dma_addr,
 				      payload | len);
-		if (!skb) {
-			cpr->sw_stats.rx.rx_oom_discards += 1;
-			rc = -ENOMEM;
-			goto next_rx;
-		}
+		if (!skb)
+			goto oom_next_rx;
 	}
 
 	if (agg_bufs) {
 		if (!xdp_active) {
 			skb = bnxt_rx_agg_pages_skb(bp, cpr, skb, cp_cons, agg_bufs, false);
-			if (!skb) {
-				cpr->sw_stats.rx.rx_oom_discards += 1;
-				rc = -ENOMEM;
-				goto next_rx;
-			}
+			if (!skb)
+				goto oom_next_rx;
 		} else {
 			skb = bnxt_xdp_build_skb(bp, skb, agg_bufs, rxr->page_pool, &xdp, rxcmp1);
 			if (!skb) {
 				/* we should be able to free the old skb here */
 				bnxt_xdp_buff_frags_free(rxr, &xdp);
-				cpr->sw_stats.rx.rx_oom_discards += 1;
-				rc = -ENOMEM;
-				goto next_rx;
+				goto oom_next_rx;
 			}
 		}
 	}
@@ -2191,6 +2178,11 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 	*raw_cons = tmp_raw_cons;
 
 	return rc;
+
+oom_next_rx:
+	cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
+	rc = -ENOMEM;
+	goto next_rx;
 }
 
 /* In netpoll mode, if we are using a combined completion ring, we need to
@@ -2237,7 +2229,7 @@ static int bnxt_force_rx_discard(struct bnxt *bp,
 	}
 	rc = bnxt_rx_pkt(bp, cpr, raw_cons, event);
 	if (rc && rc != -EBUSY)
-		cpr->sw_stats.rx.rx_netpoll_discards += 1;
+		cpr->bnapi->cp_ring.sw_stats.rx.rx_netpoll_discards += 1;
 	return rc;
 }
 
Signature · 1 line
2.43.0
author maintainer
#105 lore raw

PATCH 105/228 ARM: dts: imx6ull-tarragon: fix USB over-current polarity

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Heimpold <[email protected]>

[ Upstream commit d7f3040a565214a30e2f07dc9b91566d316e2d36 ]

Our Tarragon platform uses a active-low signal to inform the i.MX6ULL about the over-current detection.

Fixes 5e4f393ccbf0 ("ARM: dts: imx6ull: Add chargebyte Tarragon support")
---
 arch/arm/boot/dts/nxp/imx/imx6ull-tarragon-common.dtsi | 1 +
 1 file changed, 1 insertion(+)
arch/arm/boot/dts/nxp/imx/imx6ull-tarragon-common.dtsi +1 -0
index 3fdece5bd31f9..5248a058230c8 100644
@@ -805,6 +805,7 @@
 		     &pinctrl_usb_pwr>;
 	dr_mode = "host";
 	power-active-high;
+	over-current-active-low;
 	disable-over-current;
 	status = "okay";
 };
Signature · 1 line
2.43.0
author maintainer
#106 lore raw

PATCH 106/228 netfilter: nf_tables: honor table dormant flag from netdev release event path

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 8e30abc9ace4f0add4cd761dfdbfaebae5632dd2 ]

Check for table dormant flag otherwise netdev release event path tries to unregister an already unregistered hook.

[524854.857999] ------------[ cut here ]------------
[524854.858010] WARNING: CPU: 0 PID: 3386599 at net/netfilter/core.c:501 __nf_unregister_net_hook+0x21a/0x260
[...]
[524854.858848] CPU: 0 PID: 3386599 Comm: kworker/u32:2 Not tainted 6.9.0-rc3+ #365
[524854.858869] Workqueue: netns cleanup_net
[524854.858886] RIP: 0010:__nf_unregister_net_hook+0x21a/0x260
[524854.858903] Code: 24 e8 aa 73 83 ff 48 63 43 1c 83 f8 01 0f 85 3d ff ff ff e8 98 d1 f0 ff 48 8b 3c 24 e8 8f 73 83 ff 48 63 43 1c e9 26 ff ff ff <0f> 0b 48 83 c4 18 48 c7 c7 00 68 e9 82 5b 5d 41 5c 41 5d 41 5e 41
[524854.858914] RSP: 0018:ffff8881e36d79e0 EFLAGS: 00010246
[524854.858926] RAX: 0000000000000000 RBX: ffff8881339ae790 RCX: ffffffff81ba524a
[524854.858936] RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffff8881c8a16438
[524854.858945] RBP: ffff8881c8a16438 R08: 0000000000000001 R09: ffffed103c6daf34
[524854.858954] R10: ffff8881e36d79a7 R11: 0000000000000000 R12: 0000000000000005
[524854.858962] R13: ffff8881c8a16000 R14: 0000000000000000 R15: ffff8881351b5a00
[524854.858971] FS:  0000000000000000(0000) GS:ffff888390800000(0000) knlGS:0000000000000000
[524854.858982] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[524854.858991] CR2: 00007fc9be0f16f4 CR3: 00000001437cc004 CR4: 00000000001706f0
[524854.859000] Call Trace:
[524854.859006]  <TASK>
[524854.859013]  ? __warn+0x9f/0x1a0
[524854.859027]  ? __nf_unregister_net_hook+0x21a/0x260
[524854.859044]  ? report_bug+0x1b1/0x1e0
[524854.859060]  ? handle_bug+0x3c/0x70
[524854.859071]  ? exc_invalid_op+0x17/0x40
[524854.859083]  ? asm_exc_invalid_op+0x1a/0x20
[524854.859100]  ? __nf_unregister_net_hook+0x6a/0x260
[524854.859116]  ? __nf_unregister_net_hook+0x21a/0x260
[524854.859135]  nf_tables_netdev_event+0x337/0x390 [nf_tables]
[524854.859304]  ? __pfx_nf_tables_netdev_event+0x10/0x10 [nf_tables]
[524854.859461]  ? packet_notifier+0xb3/0x360
[524854.859476]  ? _raw_spin_unlock_irqrestore+0x11/0x40
[524854.859489]  ? dcbnl_netdevice_event+0x35/0x140
[524854.859507]  ? __pfx_nf_tables_netdev_event+0x10/0x10 [nf_tables]
[524854.859661]  notifier_call_chain+0x7d/0x140
[524854.859677]  unregister_netdevice_many_notify+0x5e1/0xae0
Fixes d54725cd11a5 ("netfilter: nf_tables: support for multiple devices per netdev hook")
---
 net/netfilter/nft_chain_filter.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
net/netfilter/nft_chain_filter.c +3 -1
index 274b6f7e6bb57..d170758a1eb5d 100644
@@ -338,7 +338,9 @@ static void nft_netdev_event(unsigned long event, struct net_device *dev,
 		return;
 
 	if (n > 1) {
-		nf_unregister_net_hook(ctx->net, &found->ops);
+		if (!(ctx->chain->table->flags & NFT_TABLE_F_DORMANT))
+			nf_unregister_net_hook(ctx->net, &found->ops);
+
 		list_del_rcu(&found->list);
 		kfree_rcu(found, rcu);
 		return;
Signature · 1 line
2.43.0
author maintainer
#107 lore raw

PATCH 107/228 net: phy: dp83869: Fix MII mode failure

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: MD Danish Anwar <[email protected]>

[ Upstream commit 6c9cd59dbcb09a2122b5ce0dfc07c74e6fc00dc0 ]

The DP83869 driver sets the MII bit (needed for PHY to work in MII mode) only if the op-mode is either DP83869_100M_MEDIA_CONVERT or DP83869_RGMII_100_BASE.

Some drivers i.e. ICSSG support MII mode with op-mode as DP83869_RGMII_COPPER_ETHERNET for which the MII bit is not set in dp83869 driver. As a result MII mode on ICSSG doesn't work and below log is seen.

TI DP83869 300b2400.mdio:0f: selected op-mode is not valid with MII mode icssg-prueth icssg1-eth: couldn't connect to phy ethernet-phy@0 icssg-prueth icssg1-eth: can't phy connect port MII0

Fix this by setting MII bit for DP83869_RGMII_COPPER_ETHERNET op-mode as well.

Fixes 94e86ef1b801 ("net: phy: dp83869: support mii mode when rgmii strap cfg is used")
---
 drivers/net/phy/dp83869.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/phy/dp83869.c +2 -1
index fa8c6fdcf3018..d7aaefb5226b6 100644
@@ -695,7 +695,8 @@ static int dp83869_configure_mode(struct phy_device *phydev,
 	phy_ctrl_val = dp83869->mode;
 	if (phydev->interface == PHY_INTERFACE_MODE_MII) {
 		if (dp83869->mode == DP83869_100M_MEDIA_CONVERT ||
-		    dp83869->mode == DP83869_RGMII_100_BASE) {
+		    dp83869->mode == DP83869_RGMII_100_BASE ||
+		    dp83869->mode == DP83869_RGMII_COPPER_ETHERNET) {
 			phy_ctrl_val |= DP83869_OP_MODE_MII;
 		} else {
 			phydev_err(phydev, "selected op-mode is not valid with MII mode\n");
Signature · 1 line
2.43.0
author maintainer
#108 lore raw

PATCH 108/228 net: ti: icssg-prueth: Fix signedness bug in prueth_init_rx_chns()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dan Carpenter <[email protected]>

[ Upstream commit 4dcd0e83ea1d1df9b2e0174a6d3e795b3477d64e ]

The rx_chn->irq[] array is unsigned int but it should be signed for the error handling to work. Also if k3_udma_glue_rx_get_irq() returns zero then we should return -ENXIO instead of success.

---
 drivers/net/ethernet/ti/icssg/icssg_prueth.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/net/ethernet/ti/icssg/icssg_prueth.c +5 -3
index 411898a4f38ca..4a78e8a1cabf4 100644
@@ -421,12 +421,14 @@ static int prueth_init_rx_chns(struct prueth_emac *emac,
 		if (!i)
 			fdqring_id = k3_udma_glue_rx_flow_get_fdq_id(rx_chn->rx_chn,
 								     i);
-		rx_chn->irq[i] = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i);
-		if (rx_chn->irq[i] <= 0) {
-			ret = rx_chn->irq[i];
+		ret = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i);
+		if (ret <= 0) {
+			if (!ret)
+				ret = -ENXIO;
 			netdev_err(ndev, "Failed to get rx dma irq");
 			goto fail;
 		}
+		rx_chn->irq[i] = ret;
 	}
 
 	return 0;
Signature · 1 line
2.43.0
author maintainer
#109 lore raw

PATCH 109/228 i40e: Do not use WQ_MEM_RECLAIM flag for workqueue

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sindhu Devale <[email protected]>

[ Upstream commit 2cc7d150550cc981aceedf008f5459193282425c ]

Issue reported by customer during SRIOV testing, call trace:
When both i40e and the i40iw driver are loaded, a warning in check_flush_dependency is being triggered. This seems to be because of the i40e driver workqueue is allocated with the WQ_MEM_RECLAIM flag, and the i40iw one is not.

Similar error was encountered on ice too and it was fixed by removing the flag. Do the same for i40e too.

[Feb 9 09:08] ------------[ cut here ]------------
[ +0.000004] workqueue: WQ_MEM_RECLAIM i40e:i40e_service_task [i40e] is flushing !WQ_MEM_RECLAIM infiniband:0x0
[ +0.000060] WARNING: CPU: 0 PID: 937 at kernel/workqueue.c:2966 check_flush_dependency+0x10b/0x120
[ +0.000007] Modules linked in: snd_seq_dummy snd_hrtimer snd_seq snd_timer snd_seq_device snd soundcore nls_utf8 cifs cifs_arc4 nls_ucs2_utils rdma_cm iw_cm ib_cm cifs_md4 dns_resolver netfs qrtr rfkill sunrpc vfat fat intel_rapl_msr intel_rapl_common irdma intel_uncore_frequency intel_uncore_frequency_common ice ipmi_ssif isst_if_common skx_edac nfit libnvdimm x86_pkg_temp_thermal intel_powerclamp gnss coretemp ib_uverbs rapl intel_cstate ib_core iTCO_wdt iTCO_vendor_support acpi_ipmi mei_me ipmi_si intel_uncore ioatdma i2c_i801 joydev pcspkr mei ipmi_devintf lpc_ich intel_pch_thermal i2c_smbus ipmi_msghandler acpi_power_meter acpi_pad xfs libcrc32c ast sd_mod drm_shmem_helper t10_pi drm_kms_helper sg ixgbe drm i40e ahci crct10dif_pclmul libahci crc32_pclmul igb crc32c_intel libata ghash_clmulni_intel i2c_algo_bit mdio dca wmi dm_mirror dm_region_hash dm_log dm_mod fuse
[ +0.000050] CPU: 0 PID: 937 Comm: kworker/0:3 Kdump: loaded Not tainted 6.8.0-rc2-Feb-net_dev-Qiueue-00279-gbd43c5687e05 #1 [ +0.000003] Hardware name: Intel Corporation S2600BPB/S2600BPB, BIOS SE5C620.86B.02.01.0013.121520200651 12/15/2020

[  +0.000001] Workqueue: i40e i40e_service_task [i40e]
[  +0.000024] RIP: 0010:check_flush_dependency+0x10b/0x120
[  +0.000003] Code: ff 49 8b 54 24 18 48 8d 8b b0 00 00 00 49 89 e8 48

81 c6 b0 00 00 00 48 c7 c7 b0 97 fa 9f c6 05 8a cc 1f 02 01 e8 35 b3 fd ff <0f> 0b e9 10 ff ff ff 80 3d 78 cc 1f 02 00 75 94 e9 46 ff ff ff 90 [ +0.000002] RSP: 0018:ffffbd294976bcf8 EFLAGS: 00010282
[ +0.000002] RAX: 0000000000000000 RBX: ffff94d4c483c000 RCX:
0000000000000027
[ +0.000001] RDX: ffff94d47f620bc8 RSI: 0000000000000001 RDI:
ffff94d47f620bc0
[ +0.000001] RBP: 0000000000000000 R08: 0000000000000000 R09:
00000000ffff7fff
[ +0.000001] R10: ffffbd294976bb98 R11: ffffffffa0be65e8 R12:
ffff94c5451ea180
[ +0.000001] R13: ffff94c5ab5e8000 R14: ffff94c5c20b6e05 R15:
ffff94c5f1330ab0
[ +0.000001] FS: 0000000000000000(0000) GS:ffff94d47f600000(0000) knlGS:0000000000000000
[ +0.000002] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ +0.000001] CR2: 00007f9e6f1fca70 CR3: 0000000038e20004 CR4:
00000000007706f0
[ +0.000000] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[ +0.000001] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400

[  +0.000001] PKRU: 55555554
[  +0.000001] Call Trace:
[  +0.000001]  <TASK>
[  +0.000002]  ? __warn+0x80/0x130
[  +0.000003]  ? check_flush_dependency+0x10b/0x120
[  +0.000002]  ? report_bug+0x195/0x1a0
[  +0.000005]  ? handle_bug+0x3c/0x70
[  +0.000003]  ? exc_invalid_op+0x14/0x70
[  +0.000002]  ? asm_exc_invalid_op+0x16/0x20
[  +0.000006]  ? check_flush_dependency+0x10b/0x120
[  +0.000002]  ? check_flush_dependency+0x10b/0x120
[  +0.000002]  __flush_workqueue+0x126/0x3f0
[  +0.000015]  ib_cache_cleanup_one+0x1c/0xe0 [ib_core]
[  +0.000056]  __ib_unregister_device+0x6a/0xb0 [ib_core]
[  +0.000023]  ib_unregister_device_and_put+0x34/0x50 [ib_core]
[  +0.000020]  i40iw_close+0x4b/0x90 [irdma]
[  +0.000022]  i40e_notify_client_of_netdev_close+0x54/0xc0 [i40e]
[  +0.000035]  i40e_service_task+0x126/0x190 [i40e]
[  +0.000024]  process_one_work+0x174/0x340
[  +0.000003]  worker_thread+0x27e/0x390
[  +0.000001]  ? __pfx_worker_thread+0x10/0x10
[  +0.000002]  kthread+0xdf/0x110
[  +0.000002]  ? __pfx_kthread+0x10/0x10
[  +0.000002]  ret_from_fork+0x2d/0x50
[  +0.000003]  ? __pfx_kthread+0x10/0x10
[  +0.000001]  ret_from_fork_asm+0x1b/0x30
[  +0.000004]  </TASK>
[  +0.000001] ---[ end trace 0000000000000000 ]---
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/intel/i40e/i40e_main.c +1 -1
index bb4ab3a9576b9..4c38782d95516 100644
@@ -16733,7 +16733,7 @@ static int __init i40e_init_module(void)
 	 * since we need to be able to guarantee forward progress even under
 	 * memory pressure.
 	 */
-	i40e_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0, i40e_driver_name);
+	i40e_wq = alloc_workqueue("%s", 0, 0, i40e_driver_name);
 	if (!i40e_wq) {
 		pr_err("%s: Failed to create workqueue\n", i40e_driver_name);
 		return -ENOMEM;
Signature · 1 line
2.43.0
author maintainer
#110 lore raw

PATCH 110/228 i40e: Report MFS in decimal base instead of hex

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Erwan Velu <[email protected]>

[ Upstream commit ef3c313119ea448c22da10366faa26b5b4b1a18e ]

If the MFS is set below the default (0x2600), a warning message is reported like the following :

MFS for port 1 has been set below the default: 600

This message is a bit confusing as the number shown here (600) is in fact an hexa number: 0x600 = 1536

Without any explicit "0x" prefix, this message is read like the MFS is set to 600 bytes.

MFS, as per MTUs, are usually expressed in decimal base.

This commit reports both current and default MFS values in decimal so it's less confusing for end-users.

A typical warning message looks like the following :

MFS for port 1 (1536) has been set below the default (9728)

---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/i40e/i40e_main.c +2 -2
index 4c38782d95516..f3c1df46f699c 100644
@@ -16186,8 +16186,8 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	val = FIELD_GET(I40E_PRTGL_SAH_MFS_MASK,
 			rd32(&pf->hw, I40E_PRTGL_SAH));
 	if (val < MAX_FRAME_SIZE_DEFAULT)
-		dev_warn(&pdev->dev, "MFS for port %x has been set below the default: %x\n",
-			 pf->hw.port, val);
+		dev_warn(&pdev->dev, "MFS for port %x (%d) has been set below the default (%d)\n",
+			 pf->hw.port, val, MAX_FRAME_SIZE_DEFAULT);
 
 	/* Add a filter to drop all Flow control frames from any VSI from being
 	 * transmitted. By doing so we stop a malicious VF from sending out
Signature · 1 line
2.43.0
author maintainer
#111 lore raw

PATCH 111/228 iavf: Fix TC config comparison with existing adapter TC config

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sudheer Mogilappagari <[email protected]>

[ Upstream commit 54976cf58d6168b8d15cebb395069f23b2f34b31 ]

Same number of TCs doesn't imply that underlying TC configs are same. The config could be different due to difference in number of queues in each TC. Add utility function to determine if TC configs are same.

---
 drivers/net/ethernet/intel/iavf/iavf_main.c | 30 ++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)
drivers/net/ethernet/intel/iavf/iavf_main.c +29 -1
index 335fd13e86f71..1ff381361c29d 100644
@@ -3511,6 +3511,34 @@ static void iavf_del_all_cloud_filters(struct iavf_adapter *adapter)
 	spin_unlock_bh(&adapter->cloud_filter_list_lock);
 }
 
+/**
+ * iavf_is_tc_config_same - Compare the mqprio TC config with the
+ * TC config already configured on this adapter.
+ * @adapter: board private structure
+ * @mqprio_qopt: TC config received from kernel.
+ *
+ * This function compares the TC config received from the kernel
+ * with the config already configured on the adapter.
+ *
+ * Return: True if configuration is same, false otherwise.
+ **/
+static bool iavf_is_tc_config_same(struct iavf_adapter *adapter,
+				   struct tc_mqprio_qopt *mqprio_qopt)
+{
+	struct virtchnl_channel_info *ch = &adapter->ch_config.ch_info[0];
+	int i;
+
+	if (adapter->num_tc != mqprio_qopt->num_tc)
+		return false;
+
+	for (i = 0; i < adapter->num_tc; i++) {
+		if (ch[i].count != mqprio_qopt->count[i] ||
+		    ch[i].offset != mqprio_qopt->offset[i])
+			return false;
+	}
+	return true;
+}
+
 /**
  * __iavf_setup_tc - configure multiple traffic classes
  * @netdev: network interface device structure
@@ -3568,7 +3596,7 @@ static int __iavf_setup_tc(struct net_device *netdev, void *type_data)
 		if (ret)
 			return ret;
 		/* Return if same TC config is requested */
-		if (adapter->num_tc == num_tc)
+		if (iavf_is_tc_config_same(adapter, &mqprio_qopt->qopt))
 			return 0;
 		adapter->num_tc = num_tc;
 
Signature · 1 line
2.43.0
author maintainer
#112 lore raw

PATCH 112/228 ice: fix LAG and VF lock dependency in ice_reset_vf()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jacob Keller <[email protected]>

[ Upstream commit 96fdd1f6b4ed72a741fb0eb705c0e13049b8721f ]

9f74a3dfcf83 ("ice: Fix VF Reset paths when interface in a failed over aggregate"), the ice driver has acquired the LAG mutex in ice_reset_vf(). The commit placed this lock acquisition just prior to the acquisition of the VF configuration lock.

If ice_reset_vf() acquires the configuration lock via the ICE_VF_RESET_LOCK flag, this could deadlock with ice_vc_cfg_qs_msg() because it always acquires the locks in the order of the VF configuration lock and then the LAG mutex.

Lockdep reports this violation almost immediately on creating and then removing 2 VF:

======================================================
WARNING: possible circular locking dependency detected
6.8.0-rc6 #54 Tainted: G W O
------------------------------------------------------
kworker/60:3/6771 is trying to acquire lock:
ff40d43e099380a0 (&vf->cfg_lock){+.+.}-{3:3}, at: ice_reset_vf+0x22f/0x4d0 [ice]

but task is already holding lock:
ff40d43ea1961210 (&pf->lag_mutex){+.+.}-{3:3}, at: ice_reset_vf+0xb7/0x4d0 [ice]

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (&pf->lag_mutex){+.+.}-{3:3}:

       __lock_acquire+0x4f8/0xb40
       lock_acquire+0xd4/0x2d0
       __mutex_lock+0x9b/0xbf0
       ice_vc_cfg_qs_msg+0x45/0x690 [ice]
       ice_vc_process_vf_msg+0x4f5/0x870 [ice]
       __ice_clean_ctrlq+0x2b5/0x600 [ice]
       ice_service_task+0x2c9/0x480 [ice]
       process_one_work+0x1e9/0x4d0
       worker_thread+0x1e1/0x3d0
       kthread+0x104/0x140
       ret_from_fork+0x31/0x50
       ret_from_fork_asm+0x1b/0x30

-> #0 (&vf->cfg_lock){+.+.}-{3:3}:

       check_prev_add+0xe2/0xc50
       validate_chain+0x558/0x800
       __lock_acquire+0x4f8/0xb40
       lock_acquire+0xd4/0x2d0
       __mutex_lock+0x9b/0xbf0
       ice_reset_vf+0x22f/0x4d0 [ice]
       ice_process_vflr_event+0x98/0xd0 [ice]
       ice_service_task+0x1cc/0x480 [ice]
       process_one_work+0x1e9/0x4d0
       worker_thread+0x1e1/0x3d0
       kthread+0x104/0x140
       ret_from_fork+0x31/0x50
       ret_from_fork_asm+0x1b/0x30

other info that might help us debug this:

 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock(&pf->lag_mutex);
                               lock(&vf->cfg_lock);
                               lock(&pf->lag_mutex);
  lock(&vf->cfg_lock);

*** DEADLOCK ***
4 locks held by kworker/60:3/6771:
#0: ff40d43e05428b38 ((wq_completion)ice){+.+.}-{0:0}, at: process_one_work+0x176/0x4d0 #1: ff50d06e05197e58 ((work_completion)(&pf->serv_task)){+.+.}-{0:0}, at: process_one_work+0x176/0x4d0 #2: ff40d43ea1960e50 (&pf->vfs.table_lock){+.+.}-{3:3}, at: ice_process_vflr_event+0x48/0xd0 [ice] #3: ff40d43ea1961210 (&pf->lag_mutex){+.+.}-{3:3}, at: ice_reset_vf+0xb7/0x4d0 [ice]

stack backtrace:
CPU: 60 PID: 6771 Comm: kworker/60:3 Tainted: G        W  O       6.8.0-rc6 #54
Hardware name:
Workqueue: ice ice_service_task [ice]
Call Trace:
 <TASK>
 dump_stack_lvl+0x4a/0x80
 check_noncircular+0x12d/0x150
 check_prev_add+0xe2/0xc50
 ? save_trace+0x59/0x230
 ? add_chain_cache+0x109/0x450
 validate_chain+0x558/0x800
 __lock_acquire+0x4f8/0xb40
 ? lockdep_hardirqs_on+0x7d/0x100
 lock_acquire+0xd4/0x2d0
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ? lock_is_held_type+0xc7/0x120
 __mutex_lock+0x9b/0xbf0
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ? rcu_is_watching+0x11/0x50
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ice_reset_vf+0x22f/0x4d0 [ice]
 ? process_one_work+0x176/0x4d0
 ice_process_vflr_event+0x98/0xd0 [ice]
 ice_service_task+0x1cc/0x480 [ice]
 process_one_work+0x1e9/0x4d0
 worker_thread+0x1e1/0x3d0
 ? __pfx_worker_thread+0x10/0x10
 kthread+0x104/0x140
 ? __pfx_kthread+0x10/0x10
 ret_from_fork+0x31/0x50
 ? __pfx_kthread+0x10/0x10
 ret_from_fork_asm+0x1b/0x30
 </TASK>

To avoid deadlock, we must acquire the LAG mutex only after acquiring the VF configuration lock. Fix the ice_reset_vf() to acquire the LAG mutex only after we either acquire or check that the VF configuration lock is held.

---
 drivers/net/ethernet/intel/ice/ice_vf_lib.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/net/ethernet/intel/ice/ice_vf_lib.c +8 -8
index 2ffdae9a82dfb..15ade19de55a2 100644
@@ -863,6 +863,11 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 		return 0;
 	}
 
+	if (flags & ICE_VF_RESET_LOCK)
+		mutex_lock(&vf->cfg_lock);
+	else
+		lockdep_assert_held(&vf->cfg_lock);
+
 	lag = pf->lag;
 	mutex_lock(&pf->lag_mutex);
 	if (lag && lag->bonded && lag->primary) {
@@ -874,11 +879,6 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 			act_prt = ICE_LAG_INVALID_PORT;
 	}
 
-	if (flags & ICE_VF_RESET_LOCK)
-		mutex_lock(&vf->cfg_lock);
-	else
-		lockdep_assert_held(&vf->cfg_lock);
-
 	if (ice_is_vf_disabled(vf)) {
 		vsi = ice_get_vf_vsi(vf);
 		if (!vsi) {
@@ -963,14 +963,14 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 	ice_mbx_clear_malvf(&vf->mbx_info);
 
 out_unlock:
-	if (flags & ICE_VF_RESET_LOCK)
-		mutex_unlock(&vf->cfg_lock);
-
 	if (lag && lag->bonded && lag->primary &&
 	    act_prt != ICE_LAG_INVALID_PORT)
 		ice_lag_move_vf_nodes_cfg(lag, pri_prt, act_prt);
 	mutex_unlock(&pf->lag_mutex);
 
+	if (flags & ICE_VF_RESET_LOCK)
+		mutex_unlock(&vf->cfg_lock);
+
 	return err;
 }
 
Signature · 1 line
2.43.0
author maintainer
#113 lore raw

PATCH 113/228 net: ethernet: ti: am65-cpts: Fix PTPv1 message type on TX packets

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jason Reeder <[email protected]>

[ Upstream commit 1b9e743e923b256e353a9a644195372285e5a6c0 ]

The CPTS, by design, captures the messageType (Sync, Delay_Req, etc.) field from the second nibble of the PTP header which is defined in the PTPv2 (1588-2008) specification. In the PTPv1 (1588-2002) specification the first two bytes of the PTP header are defined as the versionType which is always 0x0001. This means that any PTPv1 packets that are tagged for TX timestamping by the CPTS will have their messageType set to 0x0 which corresponds to a Sync message type. This causes issues when a PTPv1 stack is expecting a Delay_Req (messageType: 0x1) timestamp that never appears.

Fix this by checking if the ptp_class of the timestamped TX packet is PTP_CLASS_V1 and then matching the PTP sequence ID to the stored sequence ID in the skb->cb data structure. If the sequence IDs match and the packet is of type PTPv1 then there is a chance that the messageType has been incorrectly stored by the CPTS so overwrite the messageType stored by the CPTS with the messageType from the skb->cb data structure. This allows the PTPv1 stack to receive TX timestamps for Delay_Req packets which are necessary to lock onto a PTP Leader.

Fixes f6bd59526ca5 ("net: ethernet: ti: introduce am654 common platform time sync driver")
---
 drivers/net/ethernet/ti/am65-cpts.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/net/ethernet/ti/am65-cpts.c +5 -0
index c66618d91c28f..f89716b1cfb64 100644
@@ -784,6 +784,11 @@ static bool am65_cpts_match_tx_ts(struct am65_cpts *cpts,
 		struct am65_cpts_skb_cb_data *skb_cb =
 					(struct am65_cpts_skb_cb_data *)skb->cb;
 
+		if ((ptp_classify_raw(skb) & PTP_CLASS_V1) &&
+		    ((mtype_seqid & AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK) ==
+		     (skb_cb->skb_mtype_seqid & AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK)))
+			mtype_seqid = skb_cb->skb_mtype_seqid;
+
 		if (mtype_seqid == skb_cb->skb_mtype_seqid) {
 			u64 ns = event->timestamp;
 
Signature · 1 line
2.43.0
author maintainer
#114 lore raw

PATCH 114/228 octeontx2-af: fix the double free in rvu_npc_freemem()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Su Hui <[email protected]>

[ Upstream commit 6e965eba43e9724f3e603d7b7cc83e53b23d155e ]

Clang static checker(scan-build) warning: drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c:line 2184, column 2 Attempt to free released memory.

npc_mcam_rsrcs_deinit() has released 'mcam->counters.bmap'. Deleted this redundant kfree() to fix this double free problem.

Fixes dd7842878633 ("octeontx2-af: Add new devlink param to configure maximum usable NIX block LFs")
---
 drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c | 1 -
 1 file changed, 1 deletion(-)
drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c +0 -1
index 42b5ed02bc87f..d94b7b88e14ca 100644
@@ -2181,7 +2181,6 @@ void rvu_npc_freemem(struct rvu *rvu)
 
 	kfree(pkind->rsrc.bmap);
 	npc_mcam_rsrcs_deinit(rvu);
-	kfree(mcam->counters.bmap);
 	if (rvu->kpu_prfl_addr)
 		iounmap(rvu->kpu_prfl_addr);
 	else
Signature · 1 line
2.43.0
author maintainer
#115 lore raw

PATCH 115/228 dpll: check that pin is registered in __dpll_pin_unregister()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiri Pirko <[email protected]>

[ Upstream commit 9736c648370d237f61065a7e45e668e2db4374e9 ]

Similar to what is done in dpll_device_unregister(), add assertion to __dpll_pin_unregister() to make sure driver does not try to unregister non-registered pin.

Stable-dep-of: 38d7b94e81d0 ("dpll: fix dpll_pin_on_pin_register() for multiple parent pins")

---
 drivers/dpll/dpll_core.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/dpll/dpll_core.c +3 -0
index c751a87c7a8e6..3e6282937e35c 100644
@@ -29,6 +29,8 @@ static u32 dpll_pin_xa_id;
 	WARN_ON_ONCE(!xa_get_mark(&dpll_device_xa, (d)->id, DPLL_REGISTERED))
 #define ASSERT_DPLL_NOT_REGISTERED(d)	\
 	WARN_ON_ONCE(xa_get_mark(&dpll_device_xa, (d)->id, DPLL_REGISTERED))
+#define ASSERT_DPLL_PIN_REGISTERED(p) \
+	WARN_ON_ONCE(!xa_get_mark(&dpll_pin_xa, (p)->id, DPLL_REGISTERED))
 
 struct dpll_device_registration {
 	struct list_head list;
@@ -651,6 +653,7 @@ static void
 __dpll_pin_unregister(struct dpll_device *dpll, struct dpll_pin *pin,
 		      const struct dpll_pin_ops *ops, void *priv)
 {
+	ASSERT_DPLL_PIN_REGISTERED(pin);
 	dpll_xa_ref_pin_del(&dpll->pin_refs, pin, ops, priv);
 	dpll_xa_ref_dpll_del(&pin->dpll_refs, dpll, ops, priv);
 	if (xa_empty(&pin->dpll_refs))
Signature · 1 line
2.43.0
author maintainer
#116 lore raw

PATCH 116/228 dpll: fix dpll_pin_on_pin_register() for multiple parent pins

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Arkadiusz Kubalewski <[email protected]>

[ Upstream commit 38d7b94e81d068b8d8c8392f421cfd2c3bbfd1a6 ]

In scenario where pin is registered with multiple parent pins via dpll_pin_on_pin_register(..), all belonging to the same dpll device. A second call to dpll_pin_on_pin_unregister(..) would cause a call trace, as it tries to use already released registration resources (due to fix introduced in b446631f355e). In this scenario pin was registered twice, so resources are not yet expected to be release until each registered pin/pin pair is unregistered.

Currently, the following crash/call trace is produced when ice driver is removed on the system with installed E810T NIC which includes dpll device:

WARNING: CPU: 51 PID: 9155 at drivers/dpll/dpll_core.c:809 dpll_pin_ops+0x20/0x30
RIP: 0010:dpll_pin_ops+0x20/0x30
Call Trace:
 ? __warn+0x7f/0x130
 ? dpll_pin_ops+0x20/0x30
 dpll_msg_add_pin_freq+0x37/0x1d0
 dpll_cmd_pin_get_one+0x1c0/0x400
 ? __nlmsg_put+0x63/0x80
 dpll_pin_event_send+0x93/0x140
 dpll_pin_on_pin_unregister+0x3f/0x100
 ice_dpll_deinit_pins+0xa1/0x230 [ice]
 ice_remove+0xf1/0x210 [ice]

Fix by adding a parent pointer as a cookie when creating a registration, also when searching for it. For the regular pins pass NULL, this allows to create separated registration for each parent the pin is registered with.

Fixes b446631f355e ("dpll: fix dpll_xa_ref_*_del() for multiple registrations")
---
 drivers/dpll/dpll_core.c | 58 +++++++++++++++++++++++-----------------
 1 file changed, 33 insertions(+), 25 deletions(-)
drivers/dpll/dpll_core.c +33 -25
index 3e6282937e35c..a6856730ca90f 100644
@@ -42,6 +42,7 @@ struct dpll_pin_registration {
 	struct list_head list;
 	const struct dpll_pin_ops *ops;
 	void *priv;
+	void *cookie;
 };
 
 struct dpll_device *dpll_device_get_by_id(int id)
@@ -54,12 +55,14 @@ struct dpll_device *dpll_device_get_by_id(int id)
 
 static struct dpll_pin_registration *
 dpll_pin_registration_find(struct dpll_pin_ref *ref,
-			   const struct dpll_pin_ops *ops, void *priv)
+			   const struct dpll_pin_ops *ops, void *priv,
+			   void *cookie)
 {
 	struct dpll_pin_registration *reg;
 
 	list_for_each_entry(reg, &ref->registration_list, list) {
-		if (reg->ops == ops && reg->priv == priv)
+		if (reg->ops == ops && reg->priv == priv &&
+		    reg->cookie == cookie)
 			return reg;
 	}
 	return NULL;
@@ -67,7 +70,8 @@ dpll_pin_registration_find(struct dpll_pin_ref *ref,
 
 static int
 dpll_xa_ref_pin_add(struct xarray *xa_pins, struct dpll_pin *pin,
-		    const struct dpll_pin_ops *ops, void *priv)
+		    const struct dpll_pin_ops *ops, void *priv,
+		    void *cookie)
 {
 	struct dpll_pin_registration *reg;
 	struct dpll_pin_ref *ref;
@@ -78,7 +82,7 @@ dpll_xa_ref_pin_add(struct xarray *xa_pins, struct dpll_pin *pin,
 	xa_for_each(xa_pins, i, ref) {
 		if (ref->pin != pin)
 			continue;
-		reg = dpll_pin_registration_find(ref, ops, priv);
+		reg = dpll_pin_registration_find(ref, ops, priv, cookie);
 		if (reg) {
 			refcount_inc(&ref->refcount);
 			return 0;
@@ -111,6 +115,7 @@ dpll_xa_ref_pin_add(struct xarray *xa_pins, struct dpll_pin *pin,
 	}
 	reg->ops = ops;
 	reg->priv = priv;
+	reg->cookie = cookie;
 	if (ref_exists)
 		refcount_inc(&ref->refcount);
 	list_add_tail(&reg->list, &ref->registration_list);
@@ -119,7 +124,8 @@ dpll_xa_ref_pin_add(struct xarray *xa_pins, struct dpll_pin *pin,
 }
 
 static int dpll_xa_ref_pin_del(struct xarray *xa_pins, struct dpll_pin *pin,
-			       const struct dpll_pin_ops *ops, void *priv)
+			       const struct dpll_pin_ops *ops, void *priv,
+			       void *cookie)
 {
 	struct dpll_pin_registration *reg;
 	struct dpll_pin_ref *ref;
@@ -128,7 +134,7 @@ static int dpll_xa_ref_pin_del(struct xarray *xa_pins, struct dpll_pin *pin,
 	xa_for_each(xa_pins, i, ref) {
 		if (ref->pin != pin)
 			continue;
-		reg = dpll_pin_registration_find(ref, ops, priv);
+		reg = dpll_pin_registration_find(ref, ops, priv, cookie);
 		if (WARN_ON(!reg))
 			return -EINVAL;
 		list_del(&reg->list);
@@ -146,7 +152,7 @@ static int dpll_xa_ref_pin_del(struct xarray *xa_pins, struct dpll_pin *pin,
 
 static int
 dpll_xa_ref_dpll_add(struct xarray *xa_dplls, struct dpll_device *dpll,
-		     const struct dpll_pin_ops *ops, void *priv)
+		     const struct dpll_pin_ops *ops, void *priv, void *cookie)
 {
 	struct dpll_pin_registration *reg;
 	struct dpll_pin_ref *ref;
@@ -157,7 +163,7 @@ dpll_xa_ref_dpll_add(struct xarray *xa_dplls, struct dpll_device *dpll,
 	xa_for_each(xa_dplls, i, ref) {
 		if (ref->dpll != dpll)
 			continue;
-		reg = dpll_pin_registration_find(ref, ops, priv);
+		reg = dpll_pin_registration_find(ref, ops, priv, cookie);
 		if (reg) {
 			refcount_inc(&ref->refcount);
 			return 0;
@@ -190,6 +196,7 @@ dpll_xa_ref_dpll_add(struct xarray *xa_dplls, struct dpll_device *dpll,
 	}
 	reg->ops = ops;
 	reg->priv = priv;
+	reg->cookie = cookie;
 	if (ref_exists)
 		refcount_inc(&ref->refcount);
 	list_add_tail(&reg->list, &ref->registration_list);
@@ -199,7 +206,7 @@ dpll_xa_ref_dpll_add(struct xarray *xa_dplls, struct dpll_device *dpll,
 
 static void
 dpll_xa_ref_dpll_del(struct xarray *xa_dplls, struct dpll_device *dpll,
-		     const struct dpll_pin_ops *ops, void *priv)
+		     const struct dpll_pin_ops *ops, void *priv, void *cookie)
 {
 	struct dpll_pin_registration *reg;
 	struct dpll_pin_ref *ref;
@@ -208,7 +215,7 @@ dpll_xa_ref_dpll_del(struct xarray *xa_dplls, struct dpll_device *dpll,
 	xa_for_each(xa_dplls, i, ref) {
 		if (ref->dpll != dpll)
 			continue;
-		reg = dpll_pin_registration_find(ref, ops, priv);
+		reg = dpll_pin_registration_find(ref, ops, priv, cookie);
 		if (WARN_ON(!reg))
 			return;
 		list_del(&reg->list);
@@ -594,14 +601,14 @@ EXPORT_SYMBOL_GPL(dpll_pin_put);
 
 static int
 __dpll_pin_register(struct dpll_device *dpll, struct dpll_pin *pin,
-		    const struct dpll_pin_ops *ops, void *priv)
+		    const struct dpll_pin_ops *ops, void *priv, void *cookie)
 {
 	int ret;
 
-	ret = dpll_xa_ref_pin_add(&dpll->pin_refs, pin, ops, priv);
+	ret = dpll_xa_ref_pin_add(&dpll->pin_refs, pin, ops, priv, cookie);
 	if (ret)
 		return ret;
-	ret = dpll_xa_ref_dpll_add(&pin->dpll_refs, dpll, ops, priv);
+	ret = dpll_xa_ref_dpll_add(&pin->dpll_refs, dpll, ops, priv, cookie);
 	if (ret)
 		goto ref_pin_del;
 	xa_set_mark(&dpll_pin_xa, pin->id, DPLL_REGISTERED);
@@ -610,7 +617,7 @@ __dpll_pin_register(struct dpll_device *dpll, struct dpll_pin *pin,
 	return ret;
 
 ref_pin_del:
-	dpll_xa_ref_pin_del(&dpll->pin_refs, pin, ops, priv);
+	dpll_xa_ref_pin_del(&dpll->pin_refs, pin, ops, priv, cookie);
 	return ret;
 }
 
@@ -642,7 +649,7 @@ dpll_pin_register(struct dpll_device *dpll, struct dpll_pin *pin,
 		      dpll->clock_id == pin->clock_id)))
 		ret = -EINVAL;
 	else
-		ret = __dpll_pin_register(dpll, pin, ops, priv);
+		ret = __dpll_pin_register(dpll, pin, ops, priv, NULL);
 	mutex_unlock(&dpll_lock);
 
 	return ret;
@@ -651,11 +658,11 @@ EXPORT_SYMBOL_GPL(dpll_pin_register);
 
 static void
 __dpll_pin_unregister(struct dpll_device *dpll, struct dpll_pin *pin,
-		      const struct dpll_pin_ops *ops, void *priv)
+		      const struct dpll_pin_ops *ops, void *priv, void *cookie)
 {
 	ASSERT_DPLL_PIN_REGISTERED(pin);
-	dpll_xa_ref_pin_del(&dpll->pin_refs, pin, ops, priv);
-	dpll_xa_ref_dpll_del(&pin->dpll_refs, dpll, ops, priv);
+	dpll_xa_ref_pin_del(&dpll->pin_refs, pin, ops, priv, cookie);
+	dpll_xa_ref_dpll_del(&pin->dpll_refs, dpll, ops, priv, cookie);
 	if (xa_empty(&pin->dpll_refs))
 		xa_clear_mark(&dpll_pin_xa, pin->id, DPLL_REGISTERED);
 }
@@ -680,7 +687,7 @@ void dpll_pin_unregister(struct dpll_device *dpll, struct dpll_pin *pin,
 
 	mutex_lock(&dpll_lock);
 	dpll_pin_delete_ntf(pin);
-	__dpll_pin_unregister(dpll, pin, ops, priv);
+	__dpll_pin_unregister(dpll, pin, ops, priv, NULL);
 	mutex_unlock(&dpll_lock);
 }
 EXPORT_SYMBOL_GPL(dpll_pin_unregister);
@@ -716,12 +723,12 @@ int dpll_pin_on_pin_register(struct dpll_pin *parent, struct dpll_pin *pin,
 		return -EINVAL;
 
 	mutex_lock(&dpll_lock);
-	ret = dpll_xa_ref_pin_add(&pin->parent_refs, parent, ops, priv);
+	ret = dpll_xa_ref_pin_add(&pin->parent_refs, parent, ops, priv, pin);
 	if (ret)
 		goto unlock;
 	refcount_inc(&pin->refcount);
 	xa_for_each(&parent->dpll_refs, i, ref) {
-		ret = __dpll_pin_register(ref->dpll, pin, ops, priv);
+		ret = __dpll_pin_register(ref->dpll, pin, ops, priv, parent);
 		if (ret) {
 			stop = i;
 			goto dpll_unregister;
@@ -735,11 +742,12 @@ int dpll_pin_on_pin_register(struct dpll_pin *parent, struct dpll_pin *pin,
 dpll_unregister:
 	xa_for_each(&parent->dpll_refs, i, ref)
 		if (i < stop) {
-			__dpll_pin_unregister(ref->dpll, pin, ops, priv);
+			__dpll_pin_unregister(ref->dpll, pin, ops, priv,
+					      parent);
 			dpll_pin_delete_ntf(pin);
 		}
 	refcount_dec(&pin->refcount);
-	dpll_xa_ref_pin_del(&pin->parent_refs, parent, ops, priv);
+	dpll_xa_ref_pin_del(&pin->parent_refs, parent, ops, priv, pin);
 unlock:
 	mutex_unlock(&dpll_lock);
 	return ret;
@@ -764,10 +772,10 @@ void dpll_pin_on_pin_unregister(struct dpll_pin *parent, struct dpll_pin *pin,
 
 	mutex_lock(&dpll_lock);
 	dpll_pin_delete_ntf(pin);
-	dpll_xa_ref_pin_del(&pin->parent_refs, parent, ops, priv);
+	dpll_xa_ref_pin_del(&pin->parent_refs, parent, ops, priv, pin);
 	refcount_dec(&pin->refcount);
 	xa_for_each(&pin->dpll_refs, i, ref)
-		__dpll_pin_unregister(ref->dpll, pin, ops, priv);
+		__dpll_pin_unregister(ref->dpll, pin, ops, priv, parent);
 	mutex_unlock(&dpll_lock);
 }
 EXPORT_SYMBOL_GPL(dpll_pin_on_pin_unregister);
Signature · 1 line
2.43.0
author maintainer
#117 lore raw

PATCH 117/228 tls: fix lockless read of strp->msg_ready in ->poll

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sabrina Dubroca <[email protected]>

[ Upstream commit 0844370f8945086eb9335739d10205dcea8d707b ]

tls_sk_poll is called without locking the socket, and needs to read strp->msg_ready (via tls_strp_msg_ready). Convert msg_ready to a bool and use READ_ONCE/WRITE_ONCE where needed. The remaining reads are only performed when the socket is locked.

---
 include/net/tls.h  | 3 ++-
 net/tls/tls.h      | 2 +-
 net/tls/tls_strp.c | 6 +++---
 3 files changed, 6 insertions(+), 5 deletions(-)
include/net/tls.h +2 -1
index 340ad43971e47..33f657d3c0510 100644
@@ -111,7 +111,8 @@ struct tls_strparser {
 	u32 stopped : 1;
 	u32 copy_mode : 1;
 	u32 mixed_decrypted : 1;
-	u32 msg_ready : 1;
+
+	bool msg_ready;
 
 	struct strp_msg stm;
 
net/tls/tls.h +1 -1
index 762f424ff2d59..e5e47452308ab 100644
@@ -215,7 +215,7 @@ static inline struct sk_buff *tls_strp_msg(struct tls_sw_context_rx *ctx)
 
 static inline bool tls_strp_msg_ready(struct tls_sw_context_rx *ctx)
 {
-	return ctx->strp.msg_ready;
+	return READ_ONCE(ctx->strp.msg_ready);
 }
 
 static inline bool tls_strp_msg_mixed_decrypted(struct tls_sw_context_rx *ctx)
net/tls/tls_strp.c +3 -3
index ca1e0e198ceb4..5df08d848b5c9 100644
@@ -360,7 +360,7 @@ static int tls_strp_copyin(read_descriptor_t *desc, struct sk_buff *in_skb,
 	if (strp->stm.full_len && strp->stm.full_len == skb->len) {
 		desc->count = 0;
 
-		strp->msg_ready = 1;
+		WRITE_ONCE(strp->msg_ready, 1);
 		tls_rx_msg_ready(strp);
 	}
 
@@ -528,7 +528,7 @@ static int tls_strp_read_sock(struct tls_strparser *strp)
 	if (!tls_strp_check_queue_ok(strp))
 		return tls_strp_read_copy(strp, false);
 
-	strp->msg_ready = 1;
+	WRITE_ONCE(strp->msg_ready, 1);
 	tls_rx_msg_ready(strp);
 
 	return 0;
@@ -580,7 +580,7 @@ void tls_strp_msg_done(struct tls_strparser *strp)
 	else
 		tls_strp_flush_anchor_copy(strp);
 
-	strp->msg_ready = 0;
+	WRITE_ONCE(strp->msg_ready, 0);
 	memset(&strp->stm, 0, sizeof(strp->stm));
 
 	tls_strp_check_rcv(strp);
Signature · 1 line
2.43.0
author maintainer
#118 lore raw

PATCH 118/228 af_unix: Suppress false-positive lockdep splat for spin_lock() in __unix_gc().

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 1971d13ffa84a551d29a81fdf5b5ec5be166ac83 ]

syzbot reported a lockdep splat regarding unix_gc_lock and unix_state_lock().

One is called from recvmsg() for a connected socket, and another is called from GC for TCP_LISTEN socket.

So, the splat is false-positive.

Let's add a dedicated lock class for the latter to suppress the splat.

Note that this change is not necessary for net-next.git as the issue is only applied to the old GC impl.

[0]:
WARNING: possible circular locking dependency detected 6.9.0-rc5-syzkaller-00007-g4d2008430ce8 #0 Not tainted
-----------------------------------------------------
kworker/u8:1/11 is trying to acquire lock:
ffff88807cea4e70 (&u->lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline] ffff88807cea4e70 (&u->lock){+.+.}-{2:2}, at: __unix_gc+0x40e/0xf70 net/unix/garbage.c:302

but task is already holding lock:
ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline] ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: __unix_gc+0x117/0xf70 net/unix/garbage.c:261

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (unix_gc_lock){+.+.}-{2:2}:

       lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
       __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
       _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
       spin_lock include/linux/spinlock.h:351 [inline]
       unix_notinflight+0x13d/0x390 net/unix/garbage.c:140
       unix_detach_fds net/unix/af_unix.c:1819 [inline]
       unix_destruct_scm+0x221/0x350 net/unix/af_unix.c:1876
       skb_release_head_state+0x100/0x250 net/core/skbuff.c:1188
       skb_release_all net/core/skbuff.c:1200 [inline]
       __kfree_skb net/core/skbuff.c:1216 [inline]
       kfree_skb_reason+0x16d/0x3b0 net/core/skbuff.c:1252
       kfree_skb include/linux/skbuff.h:1262 [inline]
       manage_oob net/unix/af_unix.c:2672 [inline]
       unix_stream_read_generic+0x1125/0x2700 net/unix/af_unix.c:2749
       unix_stream_splice_read+0x239/0x320 net/unix/af_unix.c:2981
       do_splice_read fs/splice.c:985 [inline]
       splice_file_to_pipe+0x299/0x500 fs/splice.c:1295
       do_splice+0xf2d/0x1880 fs/splice.c:1379
       __do_splice fs/splice.c:1436 [inline]
       __do_sys_splice fs/splice.c:1652 [inline]
       __se_sys_splice+0x331/0x4a0 fs/splice.c:1634
       do_syscall_x64 arch/x86/entry/common.c:52 [inline]
       do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83
       entry_SYSCALL_64_after_hwframe+0x77/0x7f

-> #0 (&u->lock){+.+.}-{2:2}:

       check_prev_add kernel/locking/lockdep.c:3134 [inline]
       check_prevs_add kernel/locking/lockdep.c:3253 [inline]
       validate_chain+0x18cb/0x58e0 kernel/locking/lockdep.c:3869
       __lock_acquire+0x1346/0x1fd0 kernel/locking/lockdep.c:5137
       lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
       __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
       _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
       spin_lock include/linux/spinlock.h:351 [inline]
       __unix_gc+0x40e/0xf70 net/unix/garbage.c:302
       process_one_work kernel/workqueue.c:3254 [inline]
       process_scheduled_works+0xa10/0x17c0 kernel/workqueue.c:3335
       worker_thread+0x86d/0xd70 kernel/workqueue.c:3416
       kthread+0x2f0/0x390 kernel/kthread.c:388
       ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
       ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244

other info that might help us debug this:

 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock(unix_gc_lock);
                               lock(&u->lock);
                               lock(unix_gc_lock);

lock(&u->lock);

*** DEADLOCK ***

3 locks held by kworker/u8:1/11:
#0: ffff888015089148 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:3229 [inline] #0: ffff888015089148 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_scheduled_works+0x8e0/0x17c0 kernel/workqueue.c:3335 #1: ffffc90000107d00 (unix_gc_work){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:3230 [inline]
#1: ffffc90000107d00 (unix_gc_work){+.+.}-{0:0}, at: process_scheduled_works+0x91b/0x17c0 kernel/workqueue.c:3335 #2: ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline]
#2: ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: __unix_gc+0x117/0xf70 net/unix/garbage.c:261

stack backtrace:
CPU: 0 PID: 11 Comm: kworker/u8:1 Not tainted 6.9.0-rc5-syzkaller-00007-g4d2008430ce8 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024
Workqueue: events_unbound __unix_gc
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114
 check_noncircular+0x36a/0x4a0 kernel/locking/lockdep.c:2187
 check_prev_add kernel/locking/lockdep.c:3134 [inline]
 check_prevs_add kernel/locking/lockdep.c:3253 [inline]
 validate_chain+0x18cb/0x58e0 kernel/locking/lockdep.c:3869
 __lock_acquire+0x1346/0x1fd0 kernel/locking/lockdep.c:5137
 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
 __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
 _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
 spin_lock include/linux/spinlock.h:351 [inline]
 __unix_gc+0x40e/0xf70 net/unix/garbage.c:302
 process_one_work kernel/workqueue.c:3254 [inline]
 process_scheduled_works+0xa10/0x17c0 kernel/workqueue.c:3335
 worker_thread+0x86d/0xd70 kernel/workqueue.c:3416
 kthread+0x2f0/0x390 kernel/kthread.c:388
 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
 </TASK>
---
 include/net/af_unix.h | 3 +++
 net/unix/garbage.c    | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)
include/net/af_unix.h +3 -0
index d1b07ddbe677e..77bf30203d3cf 100644
@@ -77,6 +77,9 @@ enum unix_socket_lock_class {
 	U_LOCK_NORMAL,
 	U_LOCK_SECOND,	/* for double locking, see unix_state_double_lock(). */
 	U_LOCK_DIAG, /* used while dumping icons, see sk_diag_dump_icons(). */
+	U_LOCK_GC_LISTENER, /* used for listening socket while determining gc
+			     * candidates to close a small race window.
+			     */
 };
 
 static inline void unix_state_lock_nested(struct sock *sk,
net/unix/garbage.c +1 -1
index 8734c0c1fc197..2a758531e1027 100644
@@ -260,7 +260,7 @@ void unix_gc(void)
 			__set_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags);
 
 			if (sk->sk_state == TCP_LISTEN) {
-				unix_state_lock(sk);
+				unix_state_lock_nested(sk, U_LOCK_GC_LISTENER);
 				unix_state_unlock(sk);
 			}
 		}
Signature · 1 line
2.43.0
author maintainer
#119 lore raw

PATCH 119/228 netfs: Fix the pre-flush when appending to a file in writethrough mode

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Howells <[email protected]>

[ Upstream commit c97f59e276d4e93480f29a70accbd0d7273cf3f5 ]

In netfs_perform_write(), when the file is marked NETFS_ICTX_WRITETHROUGH or O_*SYNC or RWF_*SYNC was specified, write-through caching is performed on a buffered file. When setting up for write-through, we flush any conflicting writes in the region and wait for the write to complete, failing if there's a write error to return.

The issue arises if we're writing at or above the EOF position because we skip the flush and - more importantly - the wait. This becomes a problem if there's a partial folio at the end of the file that is being written out and we want to make a write to it too. Both the already-running write and the write we start both want to clear the writeback mark, but whoever is second causes a warning looking something like:

    ------------[ cut here ]------------
    R=00000012: folio 11 is not under writeback
    WARNING: CPU: 34 PID: 654 at fs/netfs/write_collect.c:105
    ...
    CPU: 34 PID: 654 Comm: kworker/u386:27 Tainted: G S ...
    ...
    Workqueue: events_unbound netfs_write_collection_worker
    ...
    RIP: 0010:netfs_writeback_lookup_folio

Fix this by making the flush-and-wait unconditional. It will do nothing if there are no folios in the pagecache and will return quickly if there are no folios in the region specified.

Further, move the WBC attachment above the flush call as the flush is going to attach a WBC and detach it again if it is not present - and since we need one anyway we might as well share it.

---
 fs/netfs/buffered_write.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)
fs/netfs/buffered_write.c +6 -7
index 8f13ca8fbc74d..267b622d923b1 100644
@@ -172,15 +172,14 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter,
 	if (unlikely(test_bit(NETFS_ICTX_WRITETHROUGH, &ctx->flags) ||
 		     iocb->ki_flags & (IOCB_DSYNC | IOCB_SYNC))
 	    ) {
-		if (pos < i_size_read(inode)) {
-			ret = filemap_write_and_wait_range(mapping, pos, pos + iter->count);
-			if (ret < 0) {
-				goto out;
-			}
-		}
-
 		wbc_attach_fdatawrite_inode(&wbc, mapping->host);
 
+		ret = filemap_write_and_wait_range(mapping, pos, pos + iter->count);
+		if (ret < 0) {
+			wbc_detach_inode(&wbc);
+			goto out;
+		}
+
 		wreq = netfs_begin_writethrough(iocb, iter->count);
 		if (IS_ERR(wreq)) {
 			wbc_detach_inode(&wbc);
Signature · 1 line
2.43.0
author maintainer
#120 lore raw

PATCH 120/228 drm/amd/display: Check DP Alt mode DPCS state via DMUB

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: George Shen <[email protected]>

[ Upstream commit 7d1e9d0369e4d45738d4b905c3ec92f76d7f91e6 ]

[Why]
Currently, driver state for DCN3.2 is not strictly matching HW state for the USBC port. To reduce inconsistencies while debugging, the driver should match HW configuration.

[How]
Update link encoder flag to indicate USBC port. Call into DMUB to check when DP Alt mode is entered, and also to check for 2-lane versuse 4-lane mode.

Stable-dep-of: 91f10a3d21f2 ("Revert "drm/amd/display: fix USB-C flag update after enc10 feature init"")

---
 .../display/dc/dcn32/dcn32_dio_link_encoder.c | 85 ++++++++++++++-----
 .../display/dc/dcn32/dcn32_dio_link_encoder.h |  5 ++
 2 files changed, 71 insertions(+), 19 deletions(-)
drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_link_encoder.c +66 -19
index d761b0df28784..e224a028d68ac 100644
@@ -34,6 +34,7 @@
 #include "dc_bios_types.h"
 #include "link_enc_cfg.h"
 
+#include "dc_dmub_srv.h"
 #include "gpio_service_interface.h"
 
 #ifndef MIN
@@ -61,6 +62,38 @@
 #define AUX_REG_WRITE(reg_name, val) \
 			dm_write_reg(CTX, AUX_REG(reg_name), val)
 
+static uint8_t phy_id_from_transmitter(enum transmitter t)
+{
+	uint8_t phy_id;
+
+	switch (t) {
+	case TRANSMITTER_UNIPHY_A:
+		phy_id = 0;
+		break;
+	case TRANSMITTER_UNIPHY_B:
+		phy_id = 1;
+		break;
+	case TRANSMITTER_UNIPHY_C:
+		phy_id = 2;
+		break;
+	case TRANSMITTER_UNIPHY_D:
+		phy_id = 3;
+		break;
+	case TRANSMITTER_UNIPHY_E:
+		phy_id = 4;
+		break;
+	case TRANSMITTER_UNIPHY_F:
+		phy_id = 5;
+		break;
+	case TRANSMITTER_UNIPHY_G:
+		phy_id = 6;
+		break;
+	default:
+		phy_id = 0;
+		break;
+	}
+	return phy_id;
+}
 
 void enc32_hw_init(struct link_encoder *enc)
 {
@@ -117,38 +150,50 @@ void dcn32_link_encoder_enable_dp_output(
 	}
 }
 
-static bool dcn32_link_encoder_is_in_alt_mode(struct link_encoder *enc)
+static bool query_dp_alt_from_dmub(struct link_encoder *enc,
+	union dmub_rb_cmd *cmd)
 {
 	struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc);
-	uint32_t dp_alt_mode_disable = 0;
-	bool is_usb_c_alt_mode = false;
 
-	if (enc->features.flags.bits.DP_IS_USB_C) {
-		/* if value == 1 alt mode is disabled, otherwise it is enabled */
-		REG_GET(RDPCSPIPE_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, &dp_alt_mode_disable);
-		is_usb_c_alt_mode = (dp_alt_mode_disable == 0);
-	}
+	memset(cmd, 0, sizeof(*cmd));
+	cmd->query_dp_alt.header.type = DMUB_CMD__VBIOS;
+	cmd->query_dp_alt.header.sub_type =
+		DMUB_CMD__VBIOS_TRANSMITTER_QUERY_DP_ALT;
+	cmd->query_dp_alt.header.payload_bytes = sizeof(cmd->query_dp_alt.data);
+	cmd->query_dp_alt.data.phy_id = phy_id_from_transmitter(enc10->base.transmitter);
+
+	if (!dc_wake_and_execute_dmub_cmd(enc->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY))
+		return false;
 
-	return is_usb_c_alt_mode;
+	return true;
 }
 
-static void dcn32_link_encoder_get_max_link_cap(struct link_encoder *enc,
+bool dcn32_link_encoder_is_in_alt_mode(struct link_encoder *enc)
+{
+	union dmub_rb_cmd cmd;
+
+	if (!query_dp_alt_from_dmub(enc, &cmd))
+		return false;
+
+	return (cmd.query_dp_alt.data.is_dp_alt_disable == 0);
+}
+
+void dcn32_link_encoder_get_max_link_cap(struct link_encoder *enc,
 	struct dc_link_settings *link_settings)
 {
-	struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc);
-	uint32_t is_in_usb_c_dp4_mode = 0;
+	union dmub_rb_cmd cmd;
 
 	dcn10_link_encoder_get_max_link_cap(enc, link_settings);
 
-	/* in usb c dp2 mode, max lane count is 2 */
-	if (enc->funcs->is_in_alt_mode && enc->funcs->is_in_alt_mode(enc)) {
-		REG_GET(RDPCSPIPE_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, &is_in_usb_c_dp4_mode);
-		if (!is_in_usb_c_dp4_mode)
-			link_settings->lane_count = MIN(LANE_COUNT_TWO, link_settings->lane_count);
-	}
+	if (!query_dp_alt_from_dmub(enc, &cmd))
+		return;
 
+	if (cmd.query_dp_alt.data.is_usb &&
+			cmd.query_dp_alt.data.is_dp4 == 0)
+		link_settings->lane_count = MIN(LANE_COUNT_TWO, link_settings->lane_count);
 }
 
+
 static const struct link_encoder_funcs dcn32_link_enc_funcs = {
 	.read_state = link_enc2_read_state,
 	.validate_output_with_stream =
@@ -203,13 +248,15 @@ void dcn32_link_encoder_construct(
 	enc10->base.hpd_source = init_data->hpd_source;
 	enc10->base.connector = init_data->connector;
 
-
 	enc10->base.preferred_engine = ENGINE_ID_UNKNOWN;
 
 	enc10->base.features = *enc_features;
 	if (enc10->base.connector.id == CONNECTOR_ID_USBC)
 		enc10->base.features.flags.bits.DP_IS_USB_C = 1;
 
+	if (enc10->base.connector.id == CONNECTOR_ID_USBC)
+		enc10->base.features.flags.bits.DP_IS_USB_C = 1;
+
 	enc10->base.transmitter = init_data->transmitter;
 
 	/* set the flag to indicate whether driver poll the I2C data pin
drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_link_encoder.h +5 -0
index bbcfce06bec01..2d5f25290ed14 100644
@@ -53,4 +53,9 @@ void dcn32_link_encoder_enable_dp_output(
 	const struct dc_link_settings *link_settings,
 	enum clock_source_id clock_source);
 
+bool dcn32_link_encoder_is_in_alt_mode(struct link_encoder *enc);
+
+void dcn32_link_encoder_get_max_link_cap(struct link_encoder *enc,
+	struct dc_link_settings *link_settings);
+
 #endif /* __DC_LINK_ENCODER__DCN32_H__ */
Signature · 1 line
2.43.0
author maintainer
#121 lore raw

PATCH 121/228 Revert "drm/amd/display: fix USB-C flag update after enc10 feature init"

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Deucher <[email protected]>

[ Upstream commit 91f10a3d21f2313485178d49efef8a3ba02bd8c7 ]

This reverts commit b5abd7f983e14054593dc91d6df2aa5f8cc67652.

This change breaks DSC on 4k monitors at 144Hz over USB-C.

---
 .../gpu/drm/amd/display/dc/dcn32/dcn32_dio_link_encoder.c | 8 +++-----
 .../gpu/drm/amd/display/dc/dcn35/dcn35_dio_link_encoder.c | 4 ++--
 2 files changed, 5 insertions(+), 7 deletions(-)
drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_link_encoder.c +3 -5
index e224a028d68ac..8a0460e863097 100644
@@ -248,14 +248,12 @@ void dcn32_link_encoder_construct(
 	enc10->base.hpd_source = init_data->hpd_source;
 	enc10->base.connector = init_data->connector;
 
-	enc10->base.preferred_engine = ENGINE_ID_UNKNOWN;
-
-	enc10->base.features = *enc_features;
 	if (enc10->base.connector.id == CONNECTOR_ID_USBC)
 		enc10->base.features.flags.bits.DP_IS_USB_C = 1;
 
-	if (enc10->base.connector.id == CONNECTOR_ID_USBC)
-		enc10->base.features.flags.bits.DP_IS_USB_C = 1;
+	enc10->base.preferred_engine = ENGINE_ID_UNKNOWN;
+
+	enc10->base.features = *enc_features;
 
 	enc10->base.transmitter = init_data->transmitter;
 
drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dio_link_encoder.c +2 -2
index 81e349d5835bb..da94e5309fbaf 100644
@@ -184,6 +184,8 @@ void dcn35_link_encoder_construct(
 	enc10->base.hpd_source = init_data->hpd_source;
 	enc10->base.connector = init_data->connector;
 
+	if (enc10->base.connector.id == CONNECTOR_ID_USBC)
+		enc10->base.features.flags.bits.DP_IS_USB_C = 1;
 
 	enc10->base.preferred_engine = ENGINE_ID_UNKNOWN;
 
@@ -238,8 +240,6 @@ void dcn35_link_encoder_construct(
 	}
 
 	enc10->base.features.flags.bits.HDMI_6GB_EN = 1;
-	if (enc10->base.connector.id == CONNECTOR_ID_USBC)
-		enc10->base.features.flags.bits.DP_IS_USB_C = 1;
 
 	if (bp_funcs->get_connector_speed_cap_info)
 		result = bp_funcs->get_connector_speed_cap_info(enc10->base.ctx->dc_bios,
Signature · 1 line
2.43.0
author maintainer
#122 lore raw

PATCH 122/228 xhci: move event processing for one interrupter to a separate function

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mathias Nyman <[email protected]>

[ Upstream commit 84ac5e4fa517f5d1da0054547a82ce905678dc08 ]

Split the main XHCI interrupt handler into a different API, so that other potential interrupters can utilize similar event ring handling. A scenario would be if a secondary interrupter required to skip pending events in the event ring, which would warrant a similar set of operations.

Stable-dep-of: 5bfc311dd6c3 ("usb: xhci: correct return value in case of STS_HCE")

---
 drivers/usb/host/xhci-ring.c | 87 +++++++++++++++++-------------------
 1 file changed, 42 insertions(+), 45 deletions(-)
drivers/usb/host/xhci-ring.c +42 -45
index 61bd29dd71a2f..2306a95d6251a 100644
@@ -3075,6 +3075,46 @@ static void xhci_clear_interrupt_pending(struct xhci_hcd *xhci,
 	}
 }
 
+static int xhci_handle_events(struct xhci_hcd *xhci, struct xhci_interrupter *ir)
+{
+	int event_loop = 0;
+	u64 temp;
+
+	xhci_clear_interrupt_pending(xhci, ir);
+
+	if (xhci->xhc_state & XHCI_STATE_DYING ||
+	    xhci->xhc_state & XHCI_STATE_HALTED) {
+		xhci_dbg(xhci, "xHCI dying, ignoring interrupt. Shouldn't IRQs be disabled?\n");
+
+		/* Clear the event handler busy flag (RW1C) */
+		temp = xhci_read_64(xhci, &ir->ir_set->erst_dequeue);
+		xhci_write_64(xhci, temp | ERST_EHB, &ir->ir_set->erst_dequeue);
+		return -ENODEV;
+	}
+
+	while (xhci_handle_event(xhci, ir) > 0) {
+		/*
+		 * If half a segment of events have been handled in one go then
+		 * update ERDP, and force isoc trbs to interrupt more often
+		 */
+		if (event_loop++ > TRBS_PER_SEGMENT / 2) {
+			xhci_update_erst_dequeue(xhci, ir, false);
+
+			if (ir->isoc_bei_interval > AVOID_BEI_INTERVAL_MIN)
+				ir->isoc_bei_interval = ir->isoc_bei_interval / 2;
+
+			event_loop = 0;
+		}
+
+		/* Update SW event ring dequeue pointer */
+		inc_deq(xhci, ir->event_ring);
+	}
+
+	xhci_update_erst_dequeue(xhci, ir, true);
+
+	return 0;
+}
+
 /*
  * xHCI spec says we can get an interrupt, and if the HC has an error condition,
  * we might get bad data out of the event ring.  Section 4.10.2.7 has a list of
@@ -3083,11 +3123,8 @@ static void xhci_clear_interrupt_pending(struct xhci_hcd *xhci,
 irqreturn_t xhci_irq(struct usb_hcd *hcd)
 {
 	struct xhci_hcd *xhci = hcd_to_xhci(hcd);
-	struct xhci_interrupter *ir;
 	irqreturn_t ret = IRQ_NONE;
-	u64 temp_64;
 	u32 status;
-	int event_loop = 0;
 
 	spin_lock(&xhci->lock);
 	/* Check if the xHC generated the interrupt, or the irq is shared */
@@ -3120,50 +3157,10 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd)
 	 */
 	status |= STS_EINT;
 	writel(status, &xhci->op_regs->status);
-
-	/* This is the handler of the primary interrupter */
-	ir = xhci->interrupters[0];
-
-	xhci_clear_interrupt_pending(xhci, ir);
-
-	if (xhci->xhc_state & XHCI_STATE_DYING ||
-	    xhci->xhc_state & XHCI_STATE_HALTED) {
-		xhci_dbg(xhci, "xHCI dying, ignoring interrupt. "
-				"Shouldn't IRQs be disabled?\n");
-		/* Clear the event handler busy flag (RW1C);
-		 * the event ring should be empty.
-		 */
-		temp_64 = xhci_read_64(xhci, &ir->ir_set->erst_dequeue);
-		xhci_write_64(xhci, temp_64 | ERST_EHB,
-				&ir->ir_set->erst_dequeue);
-		ret = IRQ_HANDLED;
-		goto out;
-	}
-
-	/* FIXME this should be a delayed service routine
-	 * that clears the EHB.
-	 */
-	while (xhci_handle_event(xhci, ir) > 0) {
-		/*
-		 * If half a segment of events have been handled in one go then
-		 * update ERDP, and force isoc trbs to interrupt more often
-		 */
-		if (event_loop++ > TRBS_PER_SEGMENT / 2) {
-			xhci_update_erst_dequeue(xhci, ir, false);
-
-			if (ir->isoc_bei_interval > AVOID_BEI_INTERVAL_MIN)
-				ir->isoc_bei_interval = ir->isoc_bei_interval / 2;
-
-			event_loop = 0;
-		}
-
-		/* Update SW event ring dequeue pointer */
-		inc_deq(xhci, ir->event_ring);
-	}
-
-	xhci_update_erst_dequeue(xhci, ir, true);
 	ret = IRQ_HANDLED;
 
+	/* This is the handler of the primary interrupter */
+	xhci_handle_events(xhci, xhci->interrupters[0]);
 out:
 	spin_unlock(&xhci->lock);
 
Signature · 1 line
2.43.0
author maintainer
#123 lore raw

PATCH 123/228 usb: xhci: correct return value in case of STS_HCE

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oliver Neukum <[email protected]>

[ Upstream commit 5bfc311dd6c376d350b39028b9000ad766ddc934 ]

If we get STS_HCE we give up on the interrupt, but for the purpose of IRQ handling that still counts as ours. We may return IRQ_NONE only if we are positive that it wasn't ours. Hence correct the default.

---
 drivers/usb/host/xhci-ring.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
drivers/usb/host/xhci-ring.c +4 -5
index 2306a95d6251a..b2868217de941 100644
@@ -3123,7 +3123,7 @@ static int xhci_handle_events(struct xhci_hcd *xhci, struct xhci_interrupter *ir
 irqreturn_t xhci_irq(struct usb_hcd *hcd)
 {
 	struct xhci_hcd *xhci = hcd_to_xhci(hcd);
-	irqreturn_t ret = IRQ_NONE;
+	irqreturn_t ret = IRQ_HANDLED;
 	u32 status;
 
 	spin_lock(&xhci->lock);
@@ -3131,12 +3131,13 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd)
 	status = readl(&xhci->op_regs->status);
 	if (status == ~(u32)0) {
 		xhci_hc_died(xhci);
-		ret = IRQ_HANDLED;
 		goto out;
 	}
 
-	if (!(status & STS_EINT))
+	if (!(status & STS_EINT)) {
+		ret = IRQ_NONE;
 		goto out;
+	}
 
 	if (status & STS_HCE) {
 		xhci_warn(xhci, "WARNING: Host Controller Error\n");
@@ -3146,7 +3147,6 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd)
 	if (status & STS_FATAL) {
 		xhci_warn(xhci, "WARNING: Host System Error\n");
 		xhci_halt(xhci);
-		ret = IRQ_HANDLED;
 		goto out;
 	}
 
@@ -3157,7 +3157,6 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd)
 	 */
 	status |= STS_EINT;
 	writel(status, &xhci->op_regs->status);
-	ret = IRQ_HANDLED;
 
 	/* This is the handler of the primary interrupter */
 	xhci_handle_events(xhci, xhci->interrupters[0]);
Signature · 1 line
2.43.0
author maintainer
#124 lore raw

PATCH 124/228 KVM: x86/pmu: Zero out PMU metadata on AMD if PMU is disabled

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

[ Upstream commit f933b88e20150f15787390e2a1754a7e412754ed ]

Move the purging of common PMU metadata from intel_pmu_refresh() to kvm_pmu_refresh(), and invoke the vendor refresh() hook if and only if the VM is supposed to have a vPMU.

KVM already denies access to the PMU based on kvm->arch.enable_pmu, as get_gp_pmc_amd() returns NULL for all PMCs in that case, i.e. KVM already violates AMD's architecture by not virtualizing a PMU (kernels have long since learned to not panic when the PMU is unavailable). But configuring the PMU as if it were enabled causes unwanted side effects, e.g. calls to kvm_pmu_trigger_event() waste an absurd number of cycles due to the all_valid_pmc_idx bitmap being non-zero.

Stable-dep-of: de120e1d692d ("KVM: x86/pmu: Set enable bits for GP counters in PERF_GLOBAL_CTRL at "RESET"")

---
 arch/x86/kvm/pmu.c           | 20 ++++++++++++++++++--
 arch/x86/kvm/vmx/pmu_intel.c | 16 ++--------------
 2 files changed, 20 insertions(+), 16 deletions(-)
arch/x86/kvm/pmu.c +18 -2
index 87cc6c8809ad8..38512954ec267 100644
@@ -741,6 +741,8 @@ static void kvm_pmu_reset(struct kvm_vcpu *vcpu)
  */
 void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 {
+	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+
 	if (KVM_BUG_ON(kvm_vcpu_has_run(vcpu), vcpu->kvm))
 		return;
 
@@ -750,8 +752,22 @@ void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 	 */
 	kvm_pmu_reset(vcpu);
 
-	bitmap_zero(vcpu_to_pmu(vcpu)->all_valid_pmc_idx, X86_PMC_IDX_MAX);
-	static_call(kvm_x86_pmu_refresh)(vcpu);
+	pmu->version = 0;
+	pmu->nr_arch_gp_counters = 0;
+	pmu->nr_arch_fixed_counters = 0;
+	pmu->counter_bitmask[KVM_PMC_GP] = 0;
+	pmu->counter_bitmask[KVM_PMC_FIXED] = 0;
+	pmu->reserved_bits = 0xffffffff00200000ull;
+	pmu->raw_event_mask = X86_RAW_EVENT_MASK;
+	pmu->global_ctrl_mask = ~0ull;
+	pmu->global_status_mask = ~0ull;
+	pmu->fixed_ctr_ctrl_mask = ~0ull;
+	pmu->pebs_enable_mask = ~0ull;
+	pmu->pebs_data_cfg_mask = ~0ull;
+	bitmap_zero(pmu->all_valid_pmc_idx, X86_PMC_IDX_MAX);
+
+	if (vcpu->kvm->arch.enable_pmu)
+		static_call(kvm_x86_pmu_refresh)(vcpu);
 }
 
 void kvm_pmu_init(struct kvm_vcpu *vcpu)
arch/x86/kvm/vmx/pmu_intel.c +2 -14
index 315c7c2ba89b1..600a021ae958b 100644
@@ -491,19 +491,6 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	u64 counter_mask;
 	int i;
 
-	pmu->nr_arch_gp_counters = 0;
-	pmu->nr_arch_fixed_counters = 0;
-	pmu->counter_bitmask[KVM_PMC_GP] = 0;
-	pmu->counter_bitmask[KVM_PMC_FIXED] = 0;
-	pmu->version = 0;
-	pmu->reserved_bits = 0xffffffff00200000ull;
-	pmu->raw_event_mask = X86_RAW_EVENT_MASK;
-	pmu->global_ctrl_mask = ~0ull;
-	pmu->global_status_mask = ~0ull;
-	pmu->fixed_ctr_ctrl_mask = ~0ull;
-	pmu->pebs_enable_mask = ~0ull;
-	pmu->pebs_data_cfg_mask = ~0ull;
-
 	memset(&lbr_desc->records, 0, sizeof(lbr_desc->records));
 
 	/*
@@ -515,8 +502,9 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		return;
 
 	entry = kvm_find_cpuid_entry(vcpu, 0xa);
-	if (!entry || !vcpu->kvm->arch.enable_pmu)
+	if (!entry)
 		return;
+
 	eax.full = entry->eax;
 	edx.full = entry->edx;
 
Signature · 1 line
2.43.0
author maintainer
#125 lore raw

PATCH 125/228 KVM: x86/pmu: Set enable bits for GP counters in PERF_GLOBAL_CTRL at "RESET"

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

[ Upstream commit de120e1d692d73c7eefa3278837b1eb68f90728a ]

Set the enable bits for general purpose counters in IA32_PERF_GLOBAL_CTRL when refreshing the PMU to emulate the MSR's architecturally defined post-RESET behavior. Per Intel's SDM:

IA32_PERF_GLOBAL_CTRL: Sets bits n-1:0 and clears the upper bits.

and

Where "n" is the number of general-purpose counters available in the processor.

AMD also documents this behavior for PerfMonV2 CPUs in one of AMD's many PPRs.

Do not set any PERF_GLOBAL_CTRL bits if there are no general purpose counters, although a literal reading of the SDM would require the CPU to set either bits 63:0 or 31:0. The intent of the behavior is to globally enable all GP counters; honor the intent, if not the letter of the law.

Leaving PERF_GLOBAL_CTRL '0' effectively breaks PMU usage in guests that haven't been updated to work with PMUs that support PERF_GLOBAL_CTRL. This bug was recently exposed when KVM added supported for AMD's PerfMonV2, i.e. when KVM started exposing a vPMU with PERF_GLOBAL_CTRL to guest software that only knew how to program v1 PMUs (that don't support PERF_GLOBAL_CTRL).

Failure to emulate the post-RESET behavior results in such guests unknowingly leaving all general purpose counters globally disabled (the entire reason the post-RESET value sets the GP counter enable bits is to maintain backwards compatibility).

The bug has likely gone unnoticed because PERF_GLOBAL_CTRL has been supported on Intel CPUs for as long as KVM has existed, i.e. hardly anyone is running guest software that isn't aware of PERF_GLOBAL_CTRL on Intel PMUs. And because up until v6.0, KVM _did_ emulate the behavior for Intel CPUs, although the old behavior was likely dumb luck.

Because (a) that old code was also broken in its own way (the history of this code is a comedy of errors), and (b) PERF_GLOBAL_CTRL was documented as having a value of '0' post-RESET in all SDMs before March 2023.

Initial vPMU support in commit f5132b01386b ("KVM: Expose a version 2 architectural PMU to a guests") *almost* got it right (again likely by dumb luck), but for some reason only set the bits if the guest PMU was advertised as v1:

        if (pmu->version == 1) {
                pmu->global_ctrl = (1 << pmu->nr_arch_gp_counters) - 1;
                return;
        }

Commit f19a0c2c2e6a ("KVM: PMU emulation: GLOBAL_CTRL MSR should be enabled on reset") then tried to remedy that goof, presumably because guest PMUs were leaving PERF_GLOBAL_CTRL '0', i.e. weren't enabling counters.

        pmu->global_ctrl = ((1 << pmu->nr_arch_gp_counters) - 1) |
                (((1ull << pmu->nr_arch_fixed_counters) - 1) << X86_PMC_IDX_FIXED);
        pmu->global_ctrl_mask = ~pmu->global_ctrl;

That was KVM's behavior up until commit c49467a45fe0 ("KVM: x86/pmu:
Don't overwrite the pmu->global_ctrl when refreshing") removed *everything*. However, it did so based on the behavior defined by the SDM , which at the time stated that "Global Perf Counter Controls" is '0' at Power-Up and RESET.

But then the March 2023 SDM (325462-079US), stealthily changed its "IA-32 and Intel 64 Processor States Following Power-up, Reset, or INIT" table to say:

IA32_PERF_GLOBAL_CTRL: Sets bits n-1:0 and clears the upper bits.

Note, kvm_pmu_refresh() can be invoked multiple times, i.e. it's not a "pure" RESET flow. But it can only be called prior to the first KVM_RUN, i.e. the guest will only ever observe the final value.

Note #2, KVM has always cleared global_ctrl during refresh (see commit f5132b01386b ("KVM: Expose a version 2 architectural PMU to a guests")), i.e. there is no danger of breaking existing setups by clobbering a value set by userspace.

---
 arch/x86/kvm/pmu.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)
arch/x86/kvm/pmu.c +14 -2
index 38512954ec267..2ab2d5213f52f 100644
@@ -766,8 +766,20 @@ void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 	pmu->pebs_data_cfg_mask = ~0ull;
 	bitmap_zero(pmu->all_valid_pmc_idx, X86_PMC_IDX_MAX);
 
-	if (vcpu->kvm->arch.enable_pmu)
-		static_call(kvm_x86_pmu_refresh)(vcpu);
+	if (!vcpu->kvm->arch.enable_pmu)
+		return;
+
+	static_call(kvm_x86_pmu_refresh)(vcpu);
+
+	/*
+	 * At RESET, both Intel and AMD CPUs set all enable bits for general
+	 * purpose counters in IA32_PERF_GLOBAL_CTRL (so that software that
+	 * was written for v1 PMUs don't unknowingly leave GP counters disabled
+	 * in the global controls).  Emulate that behavior when refreshing the
+	 * PMU so that userspace doesn't need to manually set PERF_GLOBAL_CTRL.
+	 */
+	if (kvm_pmu_has_perf_global_ctrl(pmu) && pmu->nr_arch_gp_counters)
+		pmu->global_ctrl = GENMASK_ULL(pmu->nr_arch_gp_counters - 1, 0);
 }
 
 void kvm_pmu_init(struct kvm_vcpu *vcpu)
Signature · 1 line
2.43.0
author maintainer
#126 lore raw

PATCH 126/228 drm: add drm_gem_object_is_shared_for_memory_stats() helper

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Deucher <[email protected]>

[ Upstream commit b31f5eba32ae8cc28e7cfa5a55ec8670d8c718e2 ]

Add a helper so that drm drivers can consistently report shared status via the fdinfo shared memory stats interface.

In addition to handle count, show buffers as shared if they are shared via dma-buf as well (e.g., shared with v4l or some other subsystem).

v2: switch to inline function

Stable-dep-of: a6ff969fe9cb ("drm/amdgpu: fix visible VRAM handling during faults")

---
 include/drm/drm_gem.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)
include/drm/drm_gem.h +13 -0
index 369505447acd8..2ebec3984cd44 100644
@@ -553,6 +553,19 @@ unsigned long drm_gem_lru_scan(struct drm_gem_lru *lru,
 
 int drm_gem_evict(struct drm_gem_object *obj);
 
+/**
+ * drm_gem_object_is_shared_for_memory_stats - helper for shared memory stats
+ *
+ * This helper should only be used for fdinfo shared memory stats to determine
+ * if a GEM object is shared.
+ *
+ * @obj: obj in question
+ */
+static inline bool drm_gem_object_is_shared_for_memory_stats(struct drm_gem_object *obj)
+{
+	return (obj->handle_count > 1) || obj->dma_buf;
+}
+
 #ifdef CONFIG_LOCKDEP
 /**
  * drm_gem_gpuva_set_lock() - Set the lock protecting accesses to the gpuva list.
Signature · 1 line
2.43.0
author maintainer
#127 lore raw

PATCH 127/228 drm/amdgpu: add shared fdinfo stats

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Deucher <[email protected]>

[ Upstream commit ba1a58d5b907bdf1814f8f57434aebc86233430f ]

Add shared stats. Useful for seeing shared memory.

v2: take dma-buf into account as well
v3: use the new gem helper

Stable-dep-of: a6ff969fe9cb ("drm/amdgpu: fix visible VRAM handling during faults")

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c |  4 ++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 11 +++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h |  6 ++++++
 3 files changed, 21 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c +4 -0
index 5706b282a0c7e..c7df7fa3459f1 100644
@@ -97,6 +97,10 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 		   stats.requested_visible_vram/1024UL);
 	drm_printf(p, "amd-requested-gtt:\t%llu KiB\n",
 		   stats.requested_gtt/1024UL);
+	drm_printf(p, "drm-shared-vram:\t%llu KiB\n", stats.vram_shared/1024UL);
+	drm_printf(p, "drm-shared-gtt:\t%llu KiB\n", stats.gtt_shared/1024UL);
+	drm_printf(p, "drm-shared-cpu:\t%llu KiB\n", stats.cpu_shared/1024UL);
+
 	for (hw_ip = 0; hw_ip < AMDGPU_HW_IP_NUM; ++hw_ip) {
 		if (!usage[hw_ip])
 			continue;
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +11 -0
index 425cebcc5cbff..e6f69fce539b5 100644
@@ -1276,25 +1276,36 @@ void amdgpu_bo_get_memory(struct amdgpu_bo *bo,
 			  struct amdgpu_mem_stats *stats)
 {
 	uint64_t size = amdgpu_bo_size(bo);
+	struct drm_gem_object *obj;
 	unsigned int domain;
+	bool shared;
 
 	/* Abort if the BO doesn't currently have a backing store */
 	if (!bo->tbo.resource)
 		return;
 
+	obj = &bo->tbo.base;
+	shared = drm_gem_object_is_shared_for_memory_stats(obj);
+
 	domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type);
 	switch (domain) {
 	case AMDGPU_GEM_DOMAIN_VRAM:
 		stats->vram += size;
 		if (amdgpu_bo_in_cpu_visible_vram(bo))
 			stats->visible_vram += size;
+		if (shared)
+			stats->vram_shared += size;
 		break;
 	case AMDGPU_GEM_DOMAIN_GTT:
 		stats->gtt += size;
+		if (shared)
+			stats->gtt_shared += size;
 		break;
 	case AMDGPU_GEM_DOMAIN_CPU:
 	default:
 		stats->cpu += size;
+		if (shared)
+			stats->cpu_shared += size;
 		break;
 	}
 
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +6 -0
index a3ea8a82db23a..be679c42b0b8c 100644
@@ -138,12 +138,18 @@ struct amdgpu_bo_vm {
 struct amdgpu_mem_stats {
 	/* current VRAM usage, includes visible VRAM */
 	uint64_t vram;
+	/* current shared VRAM usage, includes visible VRAM */
+	uint64_t vram_shared;
 	/* current visible VRAM usage */
 	uint64_t visible_vram;
 	/* current GTT usage */
 	uint64_t gtt;
+	/* current shared GTT usage */
+	uint64_t gtt_shared;
 	/* current system memory usage */
 	uint64_t cpu;
+	/* current shared system memory usage */
+	uint64_t cpu_shared;
 	/* sum of evicted buffers, includes visible VRAM */
 	uint64_t evicted_vram;
 	/* sum of evicted buffers due to CPU access */
Signature · 1 line
2.43.0
author maintainer
#128 lore raw

PATCH 128/228 drm/amdgpu: fix visible VRAM handling during faults

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian König <[email protected]>

[ Upstream commit a6ff969fe9cbf369e3cd0ac54261fec1122682ec ]

When we removed the hacky start code check we actually didn't took into account that *all* VRAM pages needs to be CPU accessible.

Clean up the code and unify the handling into a single helper which checks if the whole resource is CPU accessible.

The only place where a partial check would make sense is during eviction, but that is neglitible.

Fixes aed01a68047b ("drm/amdgpu: Remove TTM resource->start visible VRAM condition v2")
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 22 ++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 22 --------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    | 61 ++++++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h    |  3 ++
 5 files changed, 53 insertions(+), 57 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +1 -1
index 6adeddfb3d564..116d3756c9c35 100644
@@ -819,7 +819,7 @@ static int amdgpu_cs_bo_validate(void *param, struct amdgpu_bo *bo)
 
 	p->bytes_moved += ctx.bytes_moved;
 	if (!amdgpu_gmc_vram_full_visible(&adev->gmc) &&
-	    amdgpu_bo_in_cpu_visible_vram(bo))
+	    amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 		p->bytes_moved_vis += ctx.bytes_moved;
 
 	if (unlikely(r == -ENOMEM) && domain != bo->allowed_domains) {
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +11 -11
index e6f69fce539b5..866bfde1ca6f9 100644
@@ -620,8 +620,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 		return r;
 
 	if (!amdgpu_gmc_vram_full_visible(&adev->gmc) &&
-	    bo->tbo.resource->mem_type == TTM_PL_VRAM &&
-	    amdgpu_bo_in_cpu_visible_vram(bo))
+	    amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 		amdgpu_cs_report_moved_bytes(adev, ctx.bytes_moved,
 					     ctx.bytes_moved);
 	else
@@ -1275,23 +1274,25 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, bool evict)
 void amdgpu_bo_get_memory(struct amdgpu_bo *bo,
 			  struct amdgpu_mem_stats *stats)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+	struct ttm_resource *res = bo->tbo.resource;
 	uint64_t size = amdgpu_bo_size(bo);
 	struct drm_gem_object *obj;
 	unsigned int domain;
 	bool shared;
 
 	/* Abort if the BO doesn't currently have a backing store */
-	if (!bo->tbo.resource)
+	if (!res)
 		return;
 
 	obj = &bo->tbo.base;
 	shared = drm_gem_object_is_shared_for_memory_stats(obj);
 
-	domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type);
+	domain = amdgpu_mem_type_to_domain(res->mem_type);
 	switch (domain) {
 	case AMDGPU_GEM_DOMAIN_VRAM:
 		stats->vram += size;
-		if (amdgpu_bo_in_cpu_visible_vram(bo))
+		if (amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 			stats->visible_vram += size;
 		if (shared)
 			stats->vram_shared += size;
@@ -1392,10 +1393,7 @@ vm_fault_t amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 	/* Remember that this BO was accessed by the CPU */
 	abo->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
 
-	if (bo->resource->mem_type != TTM_PL_VRAM)
-		return 0;
-
-	if (amdgpu_bo_in_cpu_visible_vram(abo))
+	if (amdgpu_res_cpu_visible(adev, bo->resource))
 		return 0;
 
 	/* Can't move a pinned BO to visible VRAM */
@@ -1419,7 +1417,7 @@ vm_fault_t amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 
 	/* this should never happen */
 	if (bo->resource->mem_type == TTM_PL_VRAM &&
-	    !amdgpu_bo_in_cpu_visible_vram(abo))
+	    !amdgpu_res_cpu_visible(adev, bo->resource))
 		return VM_FAULT_SIGBUS;
 
 	ttm_bo_move_to_lru_tail_unlocked(bo);
@@ -1583,6 +1581,7 @@ uint32_t amdgpu_bo_get_preferred_domain(struct amdgpu_device *adev,
  */
 u64 amdgpu_bo_print_info(int id, struct amdgpu_bo *bo, struct seq_file *m)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 	struct dma_buf_attachment *attachment;
 	struct dma_buf *dma_buf;
 	const char *placement;
@@ -1591,10 +1590,11 @@ u64 amdgpu_bo_print_info(int id, struct amdgpu_bo *bo, struct seq_file *m)
 
 	if (dma_resv_trylock(bo->tbo.base.resv)) {
 		unsigned int domain;
+
 		domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type);
 		switch (domain) {
 		case AMDGPU_GEM_DOMAIN_VRAM:
-			if (amdgpu_bo_in_cpu_visible_vram(bo))
+			if (amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 				placement = "VRAM VISIBLE";
 			else
 				placement = "VRAM";
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +0 -22
index be679c42b0b8c..fa03d9e4874cc 100644
@@ -250,28 +250,6 @@ static inline u64 amdgpu_bo_mmap_offset(struct amdgpu_bo *bo)
 	return drm_vma_node_offset_addr(&bo->tbo.base.vma_node);
 }
 
-/**
- * amdgpu_bo_in_cpu_visible_vram - check if BO is (partly) in visible VRAM
- */
-static inline bool amdgpu_bo_in_cpu_visible_vram(struct amdgpu_bo *bo)
-{
-	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
-	struct amdgpu_res_cursor cursor;
-
-	if (!bo->tbo.resource || bo->tbo.resource->mem_type != TTM_PL_VRAM)
-		return false;
-
-	amdgpu_res_first(bo->tbo.resource, 0, amdgpu_bo_size(bo), &cursor);
-	while (cursor.remaining) {
-		if (cursor.start < adev->gmc.visible_vram_size)
-			return true;
-
-		amdgpu_res_next(&cursor, cursor.size);
-	}
-
-	return false;
-}
-
 /**
  * amdgpu_bo_explicit_sync - return whether the bo is explicitly synced
  */
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +38 -23

93 lines not on this page · raw message

drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +3 -0
index 65ec82141a8e0..32cf6b6f6efd9 100644
@@ -139,6 +139,9 @@ int amdgpu_vram_mgr_reserve_range(struct amdgpu_vram_mgr *mgr,
 int amdgpu_vram_mgr_query_page_status(struct amdgpu_vram_mgr *mgr,
 				      uint64_t start);
 
+bool amdgpu_res_cpu_visible(struct amdgpu_device *adev,
+			    struct ttm_resource *res);
+
 int amdgpu_ttm_init(struct amdgpu_device *adev);
 void amdgpu_ttm_fini(struct amdgpu_device *adev);
 void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev,
Signature · 1 line
2.43.0
author maintainer
#129 lore raw

PATCH 129/228 selftests/seccomp: user_notification_addfd check nextfd is available

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Terry Tritton <[email protected]>

commit 8e3c9f9f3a0742cd12b682a1766674253b33fcf0 upstream.

Currently the user_notification_addfd test checks what the next expected file descriptor will be by incrementing a variable nextfd. This does not account for file descriptors that may already be open before the test is started and will cause the test to fail if any exist.

Replace nextfd++ with a function get_next_fd which will check and return the next available file descriptor.

---
 tools/testing/selftests/seccomp/seccomp_bpf.c |   24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)
tools/testing/selftests/seccomp/seccomp_bpf.c +19 -5
@@ -4037,6 +4037,16 @@ TEST(user_notification_filter_empty_thre
 	EXPECT_GT((pollfd.revents & POLLHUP) ?: 0, 0);
 }
 
+
+int get_next_fd(int prev_fd)
+{
+	for (int i = prev_fd + 1; i < FD_SETSIZE; ++i) {
+		if (fcntl(i, F_GETFD) == -1)
+			return i;
+	}
+	_exit(EXIT_FAILURE);
+}
+
 TEST(user_notification_addfd)
 {
 	pid_t pid;
@@ -4053,7 +4063,7 @@ TEST(user_notification_addfd)
 	/* There may be arbitrary already-open fds at test start. */
 	memfd = memfd_create("test", 0);
 	ASSERT_GE(memfd, 0);
-	nextfd = memfd + 1;
+	nextfd = get_next_fd(memfd);
 
 	ret = prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
 	ASSERT_EQ(0, ret) {
@@ -4064,7 +4074,8 @@ TEST(user_notification_addfd)
 	/* Check that the basic notification machinery works */
 	listener = user_notif_syscall(__NR_getppid,
 				      SECCOMP_FILTER_FLAG_NEW_LISTENER);
-	ASSERT_EQ(listener, nextfd++);
+	ASSERT_EQ(listener, nextfd);
+	nextfd = get_next_fd(nextfd);
 
 	pid = fork();
 	ASSERT_GE(pid, 0);
@@ -4119,14 +4130,16 @@ TEST(user_notification_addfd)
 
 	/* Verify we can set an arbitrary remote fd */
 	fd = ioctl(listener, SECCOMP_IOCTL_NOTIF_ADDFD, &addfd);
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 	EXPECT_EQ(filecmp(getpid(), pid, memfd, fd), 0);
 
 	/* Verify we can set an arbitrary remote fd with large size */
 	memset(&big, 0x0, sizeof(big));
 	big.addfd = addfd;
 	fd = ioctl(listener, SECCOMP_IOCTL_NOTIF_ADDFD_BIG, &big);
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 
 	/* Verify we can set a specific remote fd */
 	addfd.newfd = 42;
@@ -4164,7 +4177,8 @@ TEST(user_notification_addfd)
 	 * Child has earlier "low" fds and now 42, so we expect the next
 	 * lowest available fd to be assigned here.
 	 */
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 	ASSERT_EQ(filecmp(getpid(), pid, memfd, fd), 0);
 
 	/*
author maintainer
#130 lore raw

PATCH 130/228 selftests/seccomp: Change the syscall used in KILL_THREAD test

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Terry Tritton <[email protected]>

commit 471dbc547612adeaa769e48498ef591c6c95a57a upstream.

The Bionic version of pthread_create used on Android calls the prctl function to give the stack and thread local storage a useful name. This will cause the KILL_THREAD test to fail as it will kill the thread as soon as it is created.

change the test to use getpid instead of prctl.

---
 tools/testing/selftests/seccomp/seccomp_bpf.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
tools/testing/selftests/seccomp/seccomp_bpf.c +4 -4
@@ -784,7 +784,7 @@ void *kill_thread(void *data)
 	bool die = (bool)data;
 
 	if (die) {
-		prctl(PR_GET_SECCOMP, 0, 0, 0, 0);
+		syscall(__NR_getpid);
 		return (void *)SIBLING_EXIT_FAILURE;
 	}
 
@@ -803,11 +803,11 @@ void kill_thread_or_group(struct __test_
 {
 	pthread_t thread;
 	void *status;
-	/* Kill only when calling __NR_prctl. */
+	/* Kill only when calling __NR_getpid. */
 	struct sock_filter filter_thread[] = {
 		BPF_STMT(BPF_LD|BPF_W|BPF_ABS,
 			offsetof(struct seccomp_data, nr)),
-		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_prctl, 0, 1),
+		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_getpid, 0, 1),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_KILL_THREAD),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW),
 	};
@@ -819,7 +819,7 @@ void kill_thread_or_group(struct __test_
 	struct sock_filter filter_process[] = {
 		BPF_STMT(BPF_LD|BPF_W|BPF_ABS,
 			offsetof(struct seccomp_data, nr)),
-		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_prctl, 0, 1),
+		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_getpid, 0, 1),
 		BPF_STMT(BPF_RET|BPF_K, kill),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW),
 	};
author maintainer
#131 lore raw

PATCH 131/228 selftests/seccomp: Handle EINVAL on unshare(CLONE_NEWPID)

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Terry Tritton <[email protected]>

commit ecaaa55c9fa5e8058445a8b891070b12208cdb6d upstream.

unshare(CLONE_NEWPID) can return EINVAL if the kernel does not have the CONFIG_PID_NS option enabled.

Add a check on these calls to skip the test if we receive EINVAL.

---
 tools/testing/selftests/seccomp/seccomp_bpf.c |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
tools/testing/selftests/seccomp/seccomp_bpf.c +8 -1
@@ -3709,7 +3709,12 @@ TEST(user_notification_sibling_pid_ns)
 	ASSERT_GE(pid, 0);
 
 	if (pid == 0) {
-		ASSERT_EQ(unshare(CLONE_NEWPID), 0);
+		ASSERT_EQ(unshare(CLONE_NEWPID), 0) {
+			if (errno == EPERM)
+				SKIP(return, "CLONE_NEWPID requires CAP_SYS_ADMIN");
+			else if (errno == EINVAL)
+				SKIP(return, "CLONE_NEWPID is invalid (missing CONFIG_PID_NS?)");
+		}
 
 		pid2 = fork();
 		ASSERT_GE(pid2, 0);
@@ -3727,6 +3732,8 @@ TEST(user_notification_sibling_pid_ns)
 	ASSERT_EQ(unshare(CLONE_NEWPID), 0) {
 		if (errno == EPERM)
 			SKIP(return, "CLONE_NEWPID requires CAP_SYS_ADMIN");
+		else if (errno == EINVAL)
+			SKIP(return, "CLONE_NEWPID is invalid (missing CONFIG_PID_NS?)");
 	}
 	ASSERT_EQ(errno, 0);
 
author maintainer
#132 lore raw

PATCH 132/228 x86/CPU/AMD: Add models 0x10-0x1f to the Zen5 range

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wenkuan Wang <[email protected]>

commit 2718a7fdf292b2dcb49c856fa8a6a955ebbbc45f upstream.

Add some more Zen5 models.

---
 arch/x86/kernel/cpu/amd.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
arch/x86/kernel/cpu/amd.c +1 -2
@@ -572,8 +572,7 @@ static void bsp_init_amd(struct cpuinfo_
 
 	case 0x1a:
 		switch (c->x86_model) {
-		case 0x00 ... 0x0f:
-		case 0x20 ... 0x2f:
+		case 0x00 ... 0x2f:
 		case 0x40 ... 0x4f:
 		case 0x70 ... 0x7f:
 			setup_force_cpu_cap(X86_FEATURE_ZEN5);
author maintainer
#133 lore raw

PATCH 133/228 x86/cpu: Fix check for RDPKRU in __show_regs()

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Kaplan <[email protected]>

commit b53c6bd5d271d023857174b8fd3e32f98ae51372 upstream.

cpu_feature_enabled(X86_FEATURE_OSPKE) does not necessarily reflect whether CR4.PKE is set on the CPU. In particular, they may differ on non-BSP CPUs before setup_pku() is executed. In this scenario, RDPKRU will #UD causing the system to hang.

Fix by checking CR4 for PKE enablement which is always correct for the current CPU.

The scenario happens by inserting a WARN* before setup_pku() in identiy_cpu() or some other diagnostic which would lead to calling __show_regs().

[ bp: Massage commit message. ]

---
 arch/x86/kernel/process_64.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/process_64.c +1 -1
@@ -138,7 +138,7 @@ void __show_regs(struct pt_regs *regs, e
 		       log_lvl, d3, d6, d7);
 	}
 
-	if (cpu_feature_enabled(X86_FEATURE_OSPKE))
+	if (cr4 & X86_CR4_PKE)
 		printk("%sPKRU: %08x\n", log_lvl, read_pkru());
 }
 
author maintainer
#134 lore raw

PATCH 134/228 rust: phy: implement `Send` for `Registration`

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wedson Almeida Filho <[email protected]>

commit df70d04d56975f527b9c965322cf56e245909071 upstream.

In preparation for requiring Send for Module implementations in the next patch.

---
 rust/kernel/net/phy.rs |    4 ++++
 1 file changed, 4 insertions(+)
rust/kernel/net/phy.rs +4 -0
@@ -640,6 +640,10 @@ pub struct Registration {
     drivers: Pin<&'static mut [DriverVTable]>,
 }
 
+// SAFETY: The only action allowed in a `Registration` instance is dropping it, which is safe to do
+// from any thread because `phy_drivers_unregister` can be called from any thread context.
+unsafe impl Send for Registration {}
+
 impl Registration {
     /// Registers a PHY driver.
     pub fn register(
author maintainer
#135 lore raw

PATCH 135/228 rust: kernel: require `Send` for `Module` implementations

In reply to Greg Kroah-Hartman

6.8-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wedson Almeida Filho <[email protected]>

commit 323617f649c0966ad5e741e47e27e06d3a680d8f upstream.

The thread that calls the module initialisation code when a module is loaded is not guaranteed [in fact, it is unlikely] to be the same one that calls the module cleanup code on module unload, therefore, Module implementations must be Send to account for them moving from one thread to another implicitly.

Cc [email protected] # 6.8.x: df70d04d5697: rust: phy: implement `Send` for `Registration`
Fixes 247b365dc8dc ("rust: add `kernel` crate")
---
 rust/kernel/lib.rs |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
rust/kernel/lib.rs +1 -1
@@ -66,7 +66,7 @@ const __LOG_PREFIX: &[u8] = b"rust_kerne
 /// The top level entrypoint to implementing a kernel module.
 ///
 /// For any teardown or cleanup operations, your type may implement [`Drop`].
-pub trait Module: Sized + Sync {
+pub trait Module: Sized + Sync + Send {
     /// Called at module initialization time.
     ///
     /// Use this method to perform whatever setup or registration your module