Skip to contentSkip to conversation

6.15.6-rc1 review

PATCH In review Greg Kroah-Hartman 14 participants 195 messages Jul 8, 2025 – Jul 10, 2025 Page 1 of 2

Subsystems (10)
lore
Share & tools
lore

Greg Kroah-Hartman

author maintainer
#0 lore raw

cover 6.15.6-rc1 review

This is the start of the stable review cycle for the 6.15.6 release. There are 178 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 10 Jul 2025 16:22:09 +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-6.15.6-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-6.15.y
and the diffstat can be found below.

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 6.15.6-rc1

Borislav Petkov (AMD) <[email protected]>
x86/process: Move the buffer clearing before MONITOR

Borislav Petkov (AMD) <[email protected]>
x86/microcode/AMD: Add TSA microcode SHAs

Borislav Petkov (AMD) <[email protected]>
KVM: SVM: Advertise TSA CPUID bits to guests

Borislav Petkov <[email protected]>

    KVM: x86: Sort CPUID_8000_0021_EAX leaf bits properly

Borislav Petkov (AMD) <[email protected]>
x86/bugs: Add a Transient Scheduler Attacks mitigation

Borislav Petkov (AMD) <[email protected]>
x86/bugs: Rename MDS machinery to something more generic

Jeongjun Park <[email protected]>
mm/vmalloc: fix data race in show_numa_info()

Andrei Kuchynski <[email protected]>
usb: typec: displayport: Fix potential deadlock

Kurt Borja <[email protected]>
platform/x86: think-lmi: Fix sysfs group cleanup

Kurt Borja <[email protected]>
platform/x86: think-lmi: Fix kobject cleanup

Kurt Borja <[email protected]>
platform/x86: think-lmi: Create ksets consecutively

Vivian Wang <[email protected]>
riscv: cpu_ops_sbi: Use static array for boot_data

Zhang Rui <[email protected]>
powercap: intel_rapl: Do not change CLAMPING bit if ENABLE bit cannot be changed

Lu Baolu <[email protected]>
iommu/vt-d: Assign devtlb cache tag on ATS enablement

Simon Xue <[email protected]>
iommu/rockchip: prevent iommus dead loop when two masters share one IOMMU

Jens Wiklander <[email protected]>
optee: ffa: fix sleep in atomic context

Oliver Neukum <[email protected]>
Logitech C-270 even more broken

Michael J. Ruhl <[email protected]>
i2c/designware: Fix an initialization issue

Krzysztof Kozlowski <[email protected]>
dt-bindings: i2c: realtek,rtl9301: Fix missing 'reg' constraint

Qasim Ijaz <[email protected]>
HID: appletb-kbd: fix slab use-after-free bug in appletb_kbd_probe

Christian König <[email protected]>
dma-buf: fix timeout handling in dma_resv_wait_timeout v2

Shyam Prasad N <[email protected]>
cifs: all initializations for tcon should happen in tcon_info_alloc

Philipp Kerling <[email protected]>
smb: client: fix readdir returning wrong type with POSIX extensions

Heikki Krogerus <[email protected]>
usb: acpi: fix device link removal

Xu Yang <[email protected]>
usb: chipidea: udc: disconnect/reconnect from host when do suspend/resume

SCHNEIDER Johannes <[email protected]>
usb: dwc3: gadget: Fix TRB reclaim logic for short transfers and ZLPs

Kuen-Han Tsai <[email protected]>
usb: dwc3: Abort suspend on soft disconnect failure

Pawel Laszczak <[email protected]>
usb: cdnsp: Fix issue with CV Bad Descriptor test

Peter Chen <[email protected]>
usb: cdnsp: do not disable slot for disabled slot

Jeff LaBundy <[email protected]>
Input: iqs7222 - explicitly define number of external channels

Nilton Perim Neto <[email protected]>
Input: xpad - support Acer NGR 200 Controller

Qasim Ijaz <[email protected]>
HID: appletb-kbd: fix memory corruption of input_handler_list

Hongyu Xie <[email protected]>
xhci: Disable stream for xHC controller with XHCI_BROKEN_STREAMS

Mathias Nyman <[email protected]>
xhci: dbc: Flush queued requests before stopping dbc

Łukasz Bartosik <[email protected]>
xhci: dbctty: disable ECHO flag by default

Raju Rangoju <[email protected]>
usb: xhci: quirk for data loss in ISOC transfers

Roy Luo <[email protected]>
Revert "usb: xhci: Implement xhci_handshake_check_state() helper"

Roy Luo <[email protected]>
usb: xhci: Skip xhci_reset in xhci_resume if xhci is being removed

Uladzislau Rezki (Sony) <[email protected]>
rcu: Return early if callback is not specified

Pablo Martin-Gomez <[email protected]>
mtd: spinand: fix memory leak of ECC engine conf

Rafael J. Wysocki <[email protected]>
ACPICA: Refuse to evaluate a method if arguments are missing

Johannes Berg <[email protected]>
wifi: ath6kl: remove WARN on bad firmware input

Johannes Berg <[email protected]>
wifi: mac80211: drop invalid source address OCB frames

Justin Sanders <[email protected]>
aoe: defer rexmit timer downdev work to workqueue

Maurizio Lombardi <[email protected]>
scsi: target: Fix NULL pointer dereference in core_scsi3_decode_spec_i_port()

Heiko Stuebner <[email protected]>
regulator: fan53555: add enable_time support and soft-start times

Raven Black <[email protected]>
ASoC: amd: yc: update quirk data for HP Victus

Madhavan Srinivasan <[email protected]>
powerpc: Fix struct termio related ioctl macros

Gyeyoung Baek <[email protected]>
genirq/irq_sim: Initialize work context pointers properly

Mario Limonciello <[email protected]>
platform/x86/amd/pmc: Add PCSpecialist Lafite Pro V 14M to 8042 quirks list

Gabriel Santese <[email protected]>
ASoC: amd: yc: Add quirk for MSI Bravo 17 D7VF internal mic

Johannes Berg <[email protected]>
ata: pata_cs5536: fix build on 32-bit UML

Tasos Sahanidis <[email protected]>
ata: libata-acpi: Do not assume 40 wire cable if no devices are enabled

Takashi Iwai <[email protected]>
ALSA: sb: Force to disable DMAs once when DMA mode is changed

Takashi Iwai <[email protected]>
ALSA: sb: Don't allow changing the DMA mode during operations

Rob Clark <[email protected]>
drm/msm: Fix another leak in the submit error path

Rob Clark <[email protected]>
drm/msm: Fix a fence leak in submit error path

Jake Hillion <[email protected]>
x86/platform/amd: move final timeout check to after final sleep

Harry Austen <[email protected]>
drm/xe: Allow dropping kunit dependency as built-in

David Howells <[email protected]>
netfs: Fix double put of request

Paulo Alcantara <[email protected]>
smb: client: fix native SMB symlink traversal

Wang Zhaolong <[email protected]>
smb: client: fix race condition in negotiate timeout by using more precise timing

Antoine Tenart <[email protected]>
net: ipv4: fix stat increase when udp early demux drops the packet

Raju Rangoju <[email protected]>
amd-xgbe: do not double read link status

Lion Ackermann <[email protected]>
net/sched: Always pass notifications when child class becomes empty

Thomas Fourier <[email protected]>
nui: Fix dma_mapping_error() check

Kohei Enju <[email protected]>
rose: fix dangling neighbour pointers in rose_rt_device_down()

Alok Tiwari <[email protected]>
enic: fix incorrect MTU comparison in enic_change_mtu()

Raju Rangoju <[email protected]>
amd-xgbe: align CL37 AN sequence as per databook

Dan Carpenter <[email protected]>
lib: test_objagg: Set error message in check_expect_hints_stats()

Vinay Belgaumkar <[email protected]>

    drm/xe/bmg: Update Wa_22019338487

Vinay Belgaumkar <[email protected]>

    drm/xe/bmg: Update Wa_14022085890

Lucas De Marchi <[email protected]>
drm/xe: Split xe_device_td_flush()

Lucas De Marchi <[email protected]>
drm/xe/guc_pc: Add _locked variant for min/max freq

John Harrison <[email protected]>

    drm/xe/guc: Enable w/a 16026508708

David Howells <[email protected]>
netfs: Fix i_size updating

Paulo Alcantara <[email protected]>
smb: client: set missing retry flag in cifs_writev_callback()

Paulo Alcantara <[email protected]>
smb: client: set missing retry flag in cifs_readv_callback()

Paulo Alcantara <[email protected]>
smb: client: set missing retry flag in smb2_writev_callback()

David Howells <[email protected]>
netfs: Fix ref leak on inserted extra subreq in write retry

David Howells <[email protected]>
netfs: Fix looping in wait functions

David Howells <[email protected]>
netfs: Fix hang due to missing case in final DIO read result collection

Jia Yao <[email protected]>
drm/xe: Fix out-of-bounds field write in MI_STORE_DATA_IMM

Vitaly Lifshits <[email protected]>
igc: disable L1.2 PCI-E link substate to avoid performance issue

Ahmed Zaki <[email protected]>
idpf: convert control queue mutex to a spinlock

Michal Swiatkowski <[email protected]>
idpf: return 0 size for RSS key if not supported

Geliang Tang <[email protected]>
nvme-multipath: fix suspicious RCU usage warning

Junxiao Chang <[email protected]>
drm/i915/gsc: mei interrupt top half should be in irq disabled context

Janusz Krzysztofik <[email protected]>
drm/i915/gt: Fix timeline left held on VMA alloc error

Marko Kiiskila <[email protected]>
drm/vmwgfx: Fix guests running with TDX/SEV

Oleksij Rempel <[email protected]>
net: usb: lan78xx: fix WARN in __netif_napi_del_locked on disconnect

Paulo Alcantara <[email protected]>
smb: client: fix warning when reconnecting channel

Dmitry Baryshkov <[email protected]>
drm/bridge: aux-hpd-bridge: fix assignment of the of_node

Dmitry Baryshkov <[email protected]>
drm/bridge: panel: move prepare_prev_first handling to drm_panel_bridge_add_typed

Alok Tiwari <[email protected]>
platform/mellanox: mlxreg-lc: Fix logic error in power state check

Kurt Borja <[email protected]>
platform/x86: dell-wmi-sysman: Fix class device unregistration

Kurt Borja <[email protected]>
platform/x86: think-lmi: Fix class device unregistration

Kurt Borja <[email protected]>
platform/x86: hp-bioscfg: Fix class device unregistration

Kurt Borja <[email protected]>
platform/x86: dell-wmi-sysman: Fix WMI data block retrieval in sysfs callbacks

Eugen Hristev <[email protected]>
nvme-pci: refresh visible attrs after being checked

Dmitry Bogdanov <[email protected]>
nvmet: fix memory leak of bio integrity

Alok Tiwari <[email protected]>
nvme: Fix incorrect cdw15 value in passthru error logging

Dan Carpenter <[email protected]>
drm/i915/selftests: Change mock_request() to return error pointers

James Clark <[email protected]>
spi: spi-fsl-dspi: Clear completion counter before initiating transfer

Gabor Juhos <[email protected]>
spi: spi-qpic-snand: reallocate BAM transactions

Marek Szyprowski <[email protected]>
drm/exynos: fimd: Guard display clock control with runtime PM calls

Fushuai Wang <[email protected]>
dpaa2-eth: fix xdp_rxq_info leak

Thomas Fourier <[email protected]>
ethernet: atl1: Add missing DMA mapping error checks and count errors

Filipe Manana <[email protected]>
btrfs: use btrfs_record_snapshot_destroy() during rmdir

Filipe Manana <[email protected]>
btrfs: propagate last_unlink_trans earlier when doing a rmdir

Filipe Manana <[email protected]>
btrfs: record new subvolume in parent dir earlier to avoid dir logging races

Filipe Manana <[email protected]>
btrfs: fix inode lookup error handling during log replay

Filipe Manana <[email protected]>
btrfs: fix iteration of extrefs during log replay

Filipe Manana <[email protected]>
btrfs: fix missing error handling when searching for inode refs during log replay

Filipe Manana <[email protected]>
btrfs: fix failure to rebuild free space tree using multiple transactions

Yang Li <[email protected]>
Bluetooth: Prevent unintended pause by checking if advertising is active

Trond Myklebust <[email protected]>
NFSv4/flexfiles: Fix handling of NFS level errors in I/O

Tigran Mkrtchyan <[email protected]>
flexfiles/pNFS: update stats on NFS4ERR_DELAY for v4.1 DSes

Armin Wolf <[email protected]>
platform/x86: wmi: Fix WMI event enablement

Alok Tiwari <[email protected]>
platform/mellanox: nvsw-sn2201: Fix bus number in adapter error message

Alok Tiwari <[email protected]>
platform/mellanox: mlxbf-pmc: Fix duplicate event ID for CACHE_DATA1

Patrisious Haddad <[email protected]>
RDMA/mlx5: Fix vport loopback for MPV device

Patrisious Haddad <[email protected]>
RDMA/mlx5: Fix CC counters query for MPV

Patrisious Haddad <[email protected]>
RDMA/mlx5: Fix HW counters query for non-representor devices

Or Har-Toov <[email protected]>
IB/mlx5: Fix potential deadlock in MR deregistration

Bart Van Assche <[email protected]>
scsi: ufs: core: Fix spelling of a sysfs attribute name

Christoph Hellwig <[email protected]>
scsi: core: Enforce unlimited max_segment_size when virt_boundary_mask is set

jackysliu <[email protected]>

    scsi: sd: Fix VPD page 0xb7 length check

Thomas Fourier <[email protected]>
scsi: qla4xxx: Fix missing DMA mapping error in qla4xxx_alloc_pdu()

Thomas Fourier <[email protected]>
scsi: qla2xxx: Fix DMA mapping test in qla24xx_get_port_database()

Benjamin Coddington <[email protected]>
NFSv4/pNFS: Fix a race to wake on NFS_LAYOUT_DRAIN

Kuniyuki Iwashima <[email protected]>
nfs: Clean up /proc/net/rpc/nfs when nfs_fs_proc_net_init() fails.

Shivank Garg <[email protected]>
fs: export anon_inode_make_secure_inode() and fix secretmem LSM bypass

Peter Zijlstra <[email protected]>
module: Provide EXPORT_SYMBOL_GPL_FOR_MODULES() helper

Mark Zhang <[email protected]>

    RDMA/mlx5: Initialize obj_event->obj_sub_list before xa_insert

Or Har-Toov <[email protected]>
RDMA/mlx5: Fix unsafe xarray access in implicit ODP handling

David Thompson <[email protected]>
platform/mellanox: mlxbf-tmfifo: fix vring_desc.len assignment

Sven Peter <[email protected]>
arm64: dts: apple: Move touchbar mipi {address,size}-cells from dtsi to dts

Sven Peter <[email protected]>
arm64: dts: apple: Drop {address,size}-cells from SPI NOR

Janne Grunau <[email protected]>

    arm64: dts: apple: t8103: Fix PCIe BCM4377 nodename

Arnd Bergmann <[email protected]>
RDMA/mlx5: reduce stack usage in mlx5_ib_ufile_hw_cleanup

Tudor Ambarus <[email protected]>
firmware: exynos-acpm: fix timeouts on xfers handling

Viresh Kumar <[email protected]>
firmware: arm_ffa: Fix the missing entry in struct ffa_indirect_msg_hdr

Sudeep Holla <[email protected]>
firmware: arm_ffa: Replace mutex with rwlock to avoid sleep in atomic context

Sudeep Holla <[email protected]>
firmware: arm_ffa: Move memory allocation outside the mutex locking

Sudeep Holla <[email protected]>
firmware: arm_ffa: Fix memory leak by freeing notifier callback node

Maíra Canal <[email protected]>
drm/v3d: Disable interrupts before resetting the GPU

Sergey Senozhatsky <[email protected]>

    mtk-sd: reset host->mrq on prepare_data() error

Masami Hiramatsu (Google) <[email protected]>
mtk-sd: Prevent memory corruption from DMA map failure

Masami Hiramatsu (Google) <[email protected]>
mtk-sd: Fix a pagefault in dma_unmap_sg() for not prepared data

RD Babiera <[email protected]>
usb: typec: altmodes/displayport: do not index invalid pin_assignments

Christian Brauner <[email protected]>
anon_inode: rework assertions

Yunshui Jiang <[email protected]>
Input: cs40l50-vibra - fix potential NULL dereference in cs40l50_upload_owt()

Manivannan Sadhasivam <[email protected]>
regulator: gpio: Fix the out-of-bounds access to drvdata::gpiods

Nicolin Chen <[email protected]>
iommufd/selftest: Fix iommufd_dirty_tracking with large hugepage sizes

Christian Eggers <[email protected]>
Bluetooth: MGMT: mesh_send: check instances prior disabling advertising

Christian Eggers <[email protected]>
Bluetooth: MGMT: set_mesh: update LE scan interval and window

Christian Eggers <[email protected]>
Bluetooth: hci_sync: revert some mesh modifications

Christian Eggers <[email protected]>
Bluetooth: HCI: Set extended advertising data synchronously

Victor Shih <[email protected]>
mmc: core: Adjust some error messages for SD UHS-II cards

Avri Altman <[email protected]>
mmc: core: sd: Apply BROKEN_SD_DISCARD quirk earlier

Victor Shih <[email protected]>
mmc: sdhci-uhs2: Adjust some error messages and register dump for SD UHS-II card

Ulf Hansson <[email protected]>
Revert "mmc: sdhci: Disable SD card clock before changing parameters"

Darrick J. Wong <[email protected]>
xfs: actually use the xfs_growfs_check_rtgeom tracepoint

Victor Shih <[email protected]>
mmc: sdhci: Add a helper function for dump register in dynamic debug mode

Jiawen Wu <[email protected]>
net: libwx: fix the incorrect display of the queue number

Nicolin Chen <[email protected]>
iommufd/selftest: Add asserts testing global mfd

Nicolin Chen <[email protected]>
iommufd/selftest: Add missing close(mfd) in memfd_mmap()

HarshaVardhana S A <[email protected]>
vsock/vmci: Clear the vmci transport packet properly when initializing it

Jiawen Wu <[email protected]>
net: txgbe: request MISC IRQ in ndo_open

Krzysztof Kozlowski <[email protected]>
dt-bindings: net: sophgo,sg2044-dwmac: Drop status from the example

Niklas Schnelle <[email protected]>
s390/pci: Do not try re-enabling load/store if device is disabled

Niklas Schnelle <[email protected]>
s390/pci: Fix stale function handles in error handling

Bui Quang Minh <[email protected]>
virtio-net: ensure the received length does not exceed allocated size

Bui Quang Minh <[email protected]>
virtio-net: xsk: rx: fix the frame's length check

Mateusz Jończyk <[email protected]>
rtc: cmos: use spin_lock_irqsave in cmos_interrupt

Elena Popa <[email protected]>
rtc: pcf2127: fix SPI command byte for PCF2131

Hugo Villeneuve <[email protected]>
rtc: pcf2127: add missing semicolon after statement

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

Diffstat:

 Documentation/ABI/testing/sysfs-devices-system-cpu |   1 +
 Documentation/ABI/testing/sysfs-driver-ufs         |   2 +-
 .../hw-vuln/processor_mmio_stale_data.rst          |   4 +-
 Documentation/admin-guide/kernel-parameters.txt    |  13 ++
 Documentation/arch/x86/mds.rst                     |   8 +-
 Documentation/core-api/symbol-namespaces.rst       |  22 ++
 .../bindings/i2c/realtek,rtl9301-i2c.yaml          |   3 +-
 .../bindings/net/sophgo,sg2044-dwmac.yaml          |   3 +-
 Makefile                                           |   4 +-
 arch/arm64/boot/dts/apple/spi1-nvram.dtsi          |   2 -
 arch/arm64/boot/dts/apple/t8103-j293.dts           |   2 +
 arch/arm64/boot/dts/apple/t8103-jxxx.dtsi          |   2 +-
 arch/arm64/boot/dts/apple/t8103.dtsi               |   2 -
 arch/arm64/boot/dts/apple/t8112-j493.dts           |   2 +
 arch/arm64/boot/dts/apple/t8112.dtsi               |   2 -
 arch/powerpc/include/uapi/asm/ioctls.h             |   8 +-
 arch/riscv/kernel/cpu_ops_sbi.c                    |   6 +-
 arch/s390/pci/pci_event.c                          |  15 ++
 arch/x86/Kconfig                                   |   9 +
 arch/x86/entry/entry.S                             |   8 +-
 arch/x86/include/asm/cpufeatures.h                 |   5 +
 arch/x86/include/asm/irqflags.h                    |   4 +-
 arch/x86/include/asm/kvm_host.h                    |   1 +
 arch/x86/include/asm/mwait.h                       |  28 ++-
 arch/x86/include/asm/nospec-branch.h               |  37 +--
 arch/x86/kernel/cpu/amd.c                          |  44 ++++
 arch/x86/kernel/cpu/bugs.c                         | 133 ++++++++++-
 arch/x86/kernel/cpu/common.c                       |  14 +-
 arch/x86/kernel/cpu/microcode/amd_shas.c           | 112 +++++++++
 arch/x86/kernel/cpu/scattered.c                    |   2 +
 arch/x86/kernel/process.c                          |  16 +-
 arch/x86/kvm/cpuid.c                               |  15 +-
 arch/x86/kvm/reverse_cpuid.h                       |   7 +
 arch/x86/kvm/svm/vmenter.S                         |   6 +
 arch/x86/kvm/vmx/vmx.c                             |   2 +-
 drivers/acpi/acpica/dsmethod.c                     |   7 +
 drivers/ata/libata-acpi.c                          |  24 +-
 drivers/ata/pata_cs5536.c                          |   2 +-
 drivers/ata/pata_via.c                             |   6 +-
 drivers/base/cpu.c                                 |   3 +
 drivers/block/aoe/aoe.h                            |   1 +
 drivers/block/aoe/aoecmd.c                         |   8 +-
 drivers/block/aoe/aoedev.c                         |   5 +-
 drivers/dma-buf/dma-resv.c                         |  12 +-
 drivers/firmware/arm_ffa/driver.c                  |  71 +++---
 drivers/firmware/samsung/exynos-acpm.c             |  27 +--
 drivers/gpu/drm/bridge/aux-hpd-bridge.c            |   3 +-
 drivers/gpu/drm/bridge/panel.c                     |   5 +-
 drivers/gpu/drm/exynos/exynos_drm_fimd.c           |  12 +
 drivers/gpu/drm/i915/gt/intel_gsc.c                |   2 +-
 drivers/gpu/drm/i915/gt/intel_ring_submission.c    |   3 +-
 drivers/gpu/drm/i915/selftests/i915_request.c      |  20 +-
 drivers/gpu/drm/i915/selftests/mock_request.c      |   2 +-
 drivers/gpu/drm/msm/msm_gem_submit.c               |  17 +-
 drivers/gpu/drm/v3d/v3d_drv.h                      |   8 +
 drivers/gpu/drm/v3d/v3d_gem.c                      |   2 +
 drivers/gpu/drm/v3d/v3d_irq.c                      |  37 ++-
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c                |   2 +-
 drivers/gpu/drm/xe/Kconfig                         |   3 +-
 drivers/gpu/drm/xe/abi/guc_klvs_abi.h              |   1 +
 drivers/gpu/drm/xe/xe_device.c                     |  72 +++---
 drivers/gpu/drm/xe/xe_guc_ads.c                    |   5 +
 drivers/gpu/drm/xe/xe_guc_pc.c                     | 249 ++++++++++++++++-----
 drivers/gpu/drm/xe/xe_guc_pc.h                     |   2 +
 drivers/gpu/drm/xe/xe_guc_pc_types.h               |   2 +
 drivers/gpu/drm/xe/xe_migrate.c                    |  18 +-
 drivers/gpu/drm/xe/xe_wa_oob.rules                 |   6 +
 drivers/hid/hid-appletb-kbd.c                      |  14 +-
 drivers/i2c/busses/i2c-designware-master.c         |   1 +
 drivers/infiniband/hw/mlx5/counters.c              |   4 +-
 drivers/infiniband/hw/mlx5/devx.c                  |  10 +-
 drivers/infiniband/hw/mlx5/main.c                  |  33 +++
 drivers/infiniband/hw/mlx5/mr.c                    |  61 +++--
 drivers/infiniband/hw/mlx5/odp.c                   |   8 +-
 drivers/input/joystick/xpad.c                      |   2 +
 drivers/input/misc/cs40l50-vibra.c                 |   2 +
 drivers/input/misc/iqs7222.c                       |   7 +-
 drivers/iommu/intel/cache.c                        |   5 +-
 drivers/iommu/intel/iommu.c                        |  11 +-
 drivers/iommu/intel/iommu.h                        |   2 +
 drivers/iommu/rockchip-iommu.c                     |   3 +-
 drivers/mmc/core/quirks.h                          |  12 +-
 drivers/mmc/core/sd_uhs2.c                         |   4 +-
 drivers/mmc/host/mtk-sd.c                          |  21 +-
 drivers/mmc/host/sdhci-uhs2.c                      |  20 +-
 drivers/mmc/host/sdhci.c                           |   9 +-
 drivers/mmc/host/sdhci.h                           |  16 ++
 drivers/mtd/nand/spi/core.c                        |   1 +
 drivers/net/ethernet/amd/xgbe/xgbe-common.h        |   2 +
 drivers/net/ethernet/amd/xgbe/xgbe-mdio.c          |  13 ++
 drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c        |  24 +-
 drivers/net/ethernet/amd/xgbe/xgbe.h               |   4 +-
 drivers/net/ethernet/atheros/atlx/atl1.c           |  79 +++++--
 drivers/net/ethernet/cisco/enic/enic_main.c        |   4 +-
 drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c   |  26 ++-
 drivers/net/ethernet/intel/idpf/idpf_controlq.c    |  23 +-
 .../net/ethernet/intel/idpf/idpf_controlq_api.h    |   2 +-
 drivers/net/ethernet/intel/idpf/idpf_ethtool.c     |   4 +-
 drivers/net/ethernet/intel/idpf/idpf_lib.c         |  12 +-
 drivers/net/ethernet/intel/igc/igc_main.c          |  10 +
 drivers/net/ethernet/sun/niu.c                     |  31 ++-
 drivers/net/ethernet/sun/niu.h                     |   4 +
 drivers/net/ethernet/wangxun/libwx/wx_lib.c        |   1 +
 drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c     |   2 +-
 drivers/net/ethernet/wangxun/txgbe/txgbe_main.c    |  22 +-
 drivers/net/usb/lan78xx.c                          |   2 -
 drivers/net/virtio_net.c                           |  60 ++++-
 drivers/net/wireless/ath/ath6kl/bmi.c              |   4 +-
 drivers/nvme/host/core.c                           |   2 +-
 drivers/nvme/host/multipath.c                      |   3 +-
 drivers/nvme/host/pci.c                            |   6 +-
 drivers/nvme/target/nvmet.h                        |   2 +
 drivers/platform/mellanox/mlxbf-pmc.c              |   2 +-
 drivers/platform/mellanox/mlxbf-tmfifo.c           |   3 +-
 drivers/platform/mellanox/mlxreg-lc.c              |   2 +-
 drivers/platform/mellanox/nvsw-sn2201.c            |   2 +-
 drivers/platform/x86/amd/hsmp/hsmp.c               |   6 +-
 drivers/platform/x86/amd/pmc/pmc-quirks.c          |   9 +
 .../x86/dell/dell-wmi-sysman/dell-wmi-sysman.h     |   5 +
 .../x86/dell/dell-wmi-sysman/enum-attributes.c     |   5 +-
 .../x86/dell/dell-wmi-sysman/int-attributes.c      |   5 +-
 .../x86/dell/dell-wmi-sysman/passobj-attributes.c  |   5 +-
 .../x86/dell/dell-wmi-sysman/string-attributes.c   |   5 +-
 drivers/platform/x86/dell/dell-wmi-sysman/sysman.c |  12 +-
 drivers/platform/x86/hp/hp-bioscfg/bioscfg.c       |   4 +-
 drivers/platform/x86/think-lmi.c                   |  94 +++-----
 drivers/platform/x86/wmi.c                         |  16 +-
 drivers/powercap/intel_rapl_common.c               |  18 +-
 drivers/regulator/fan53555.c                       |  14 ++
 drivers/regulator/gpio-regulator.c                 |   8 +-
 drivers/rtc/rtc-cmos.c                             |  10 +-
 drivers/rtc/rtc-pcf2127.c                          |   7 +-
 drivers/scsi/hosts.c                               |  18 +-
 drivers/scsi/qla2xxx/qla_mbx.c                     |   2 +-
 drivers/scsi/qla4xxx/ql4_os.c                      |   2 +
 drivers/scsi/sd.c                                  |   2 +-
 drivers/spi/spi-fsl-dspi.c                         |  11 +-
 drivers/spi/spi-qpic-snand.c                       |  16 ++
 drivers/target/target_core_pr.c                    |   4 +-
 drivers/tee/optee/ffa_abi.c                        |  41 +++-
 drivers/tee/optee/optee_private.h                  |   2 +
 drivers/ufs/core/ufs-sysfs.c                       |   4 +-
 drivers/usb/cdns3/cdnsp-debug.h                    |   5 +-
 drivers/usb/cdns3/cdnsp-ep0.c                      |  18 +-
 drivers/usb/cdns3/cdnsp-gadget.h                   |   6 +
 drivers/usb/cdns3/cdnsp-ring.c                     |   7 +-
 drivers/usb/chipidea/udc.c                         |   7 +
 drivers/usb/core/hub.c                             |   3 +
 drivers/usb/core/quirks.c                          |   3 +-
 drivers/usb/core/usb-acpi.c                        |   4 +-
 drivers/usb/dwc3/core.c                            |   9 +-
 drivers/usb/dwc3/gadget.c                          |  24 +-
 drivers/usb/host/xhci-dbgcap.c                     |   4 +
 drivers/usb/host/xhci-dbgtty.c                     |   1 +
 drivers/usb/host/xhci-mem.c                        |   4 +
 drivers/usb/host/xhci-pci.c                        |  25 +++
 drivers/usb/host/xhci-plat.c                       |   3 +-
 drivers/usb/host/xhci-ring.c                       |   5 +-
 drivers/usb/host/xhci.c                            |  31 +--
 drivers/usb/host/xhci.h                            |   3 +-
 drivers/usb/typec/altmodes/displayport.c           |   5 +-
 fs/anon_inodes.c                                   |  23 +-
 fs/btrfs/block-group.h                             |   2 +
 fs/btrfs/free-space-tree.c                         |  40 ++++
 fs/btrfs/inode.c                                   |  36 +--
 fs/btrfs/ioctl.c                                   |   4 +-
 fs/btrfs/tree-log.c                                | 137 ++++++------
 fs/exec.c                                          |   9 +-
 fs/libfs.c                                         |   8 +-
 fs/namei.c                                         |   2 +-
 fs/netfs/buffered_write.c                          |   2 +
 fs/netfs/direct_write.c                            |   8 +-
 fs/netfs/misc.c                                    |  26 ++-
 fs/netfs/write_retry.c                             |   2 +-
 fs/nfs/flexfilelayout/flexfilelayout.c             | 120 +++++++---
 fs/nfs/inode.c                                     |  17 +-
 fs/nfs/pnfs.c                                      |   4 +-
 fs/smb/client/cifsglob.h                           |   2 +
 fs/smb/client/cifsproto.h                          |   1 +
 fs/smb/client/cifssmb.c                            |   2 +
 fs/smb/client/connect.c                            |  15 +-
 fs/smb/client/fs_context.c                         |  17 +-
 fs/smb/client/misc.c                               |   6 +
 fs/smb/client/readdir.c                            |   2 +-
 fs/smb/client/reparse.c                            |  22 +-
 fs/smb/client/smb2pdu.c                            |  11 +-
 fs/xfs/xfs_rtalloc.c                               |   2 +
 include/linux/arm_ffa.h                            |   1 +
 include/linux/cpu.h                                |   1 +
 include/linux/export.h                             |  12 +-
 include/linux/fs.h                                 |   2 +
 include/linux/libata.h                             |   7 +-
 include/linux/usb.h                                |   2 +
 include/linux/usb/typec_dp.h                       |   1 +
 include/trace/events/netfs.h                       |   1 +
 kernel/irq/irq_sim.c                               |   2 +-
 kernel/rcu/tree.c                                  |   4 +
 lib/test_objagg.c                                  |   4 +-
 mm/secretmem.c                                     |   9 +-
 mm/vmalloc.c                                       |  63 +++---
 net/bluetooth/hci_event.c                          |  36 ---
 net/bluetooth/hci_sync.c                           | 227 +++++++++++--------
 net/bluetooth/mgmt.c                               |  25 ++-
 net/ipv4/ip_input.c                                |   7 +-
 net/mac80211/rx.c                                  |   4 +
 net/rose/rose_route.c                              |  15 +-
 net/sched/sch_api.c                                |  19 +-
 net/vmw_vsock/vmci_transport.c                     |   4 +-
 sound/isa/sb/sb16_main.c                           |   7 +
 sound/soc/amd/yc/acp6x-mach.c                      |  14 ++
 tools/testing/selftests/iommu/iommufd.c            |  32 ++-
 tools/testing/selftests/iommu/iommufd_utils.h      |   9 +-
 212 files changed, 2312 insertions(+), 986 deletions(-)
author maintainer
#1 lore raw

PATCH 1/178 rtc: pcf2127: add missing semicolon after statement

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

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

From: Hugo Villeneuve <[email protected]>

commit 08d82d0cad51c2b1d454fe41ea1ff96ade676961 upstream.

Replace comma with semicolon at the end of the statement when setting config.max_register.

Fixes fd28ceb4603f ("rtc: pcf2127: add variant-specific configuration structure")
Cc Elena Popa <[email protected]>
---
 drivers/rtc/rtc-pcf2127.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/rtc/rtc-pcf2127.c +1 -1
@@ -1538,7 +1538,7 @@ static int pcf2127_spi_probe(struct spi_
 		variant = &pcf21xx_cfg[type];
 	}
 
-	config.max_register = variant->max_register,
+	config.max_register = variant->max_register;
 
 	regmap = devm_regmap_init_spi(spi, &config);
 	if (IS_ERR(regmap)) {
author maintainer
#2 lore raw

PATCH 2/178 rtc: pcf2127: fix SPI command byte for PCF2131

In reply to Greg Kroah-Hartman

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

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

From: Elena Popa <[email protected]>

commit fa78e9b606a472495ef5b6b3d8b45c37f7727f9d upstream.

PCF2131 was not responding to read/write operations using SPI. PCF2131 has a different command byte definition, compared to PCF2127/29. Added the new command byte definition when PCF2131 is detected.

---
 drivers/rtc/rtc-pcf2127.c |    5 +++++
 1 file changed, 5 insertions(+)
drivers/rtc/rtc-pcf2127.c +5 -0
@@ -1465,6 +1465,11 @@ static int pcf2127_i2c_probe(struct i2c_
 		variant = &pcf21xx_cfg[type];
 	}
 
+	if (variant->type == PCF2131) {
+		config.read_flag_mask = 0x0;
+		config.write_flag_mask = 0x0;
+	}
+
 	config.max_register = variant->max_register,
 
 	regmap = devm_regmap_init(&client->dev, &pcf2127_i2c_regmap,
author maintainer
#3 lore raw

PATCH 3/178 rtc: cmos: use spin_lock_irqsave in cmos_interrupt

In reply to Greg Kroah-Hartman

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

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

From: Mateusz Jończyk <[email protected]>

commit 00a39d8652ff9088de07a6fe6e9e1893452fe0dd upstream.

cmos_interrupt() can be called in a non-interrupt context, such as in an ACPI event handler (which runs in an interrupt thread). Therefore, usage of spin_lock(&rtc_lock) is insecure. Use spin_lock_irqsave() / spin_unlock_irqrestore() instead.

Before a misguided
commit 6950d046eb6e ("rtc: cmos: Replace spin_lock_irqsave with spin_lock in hard IRQ") the cmos_interrupt() function used spin_lock_irqsave(). That commit
changed it to spin_lock() and broke locking, which was partially fixed in commit 13be2efc390a ("rtc: cmos: Disable irq around direct invocation of cmos_interrupt()")

That second commit did not take account of the ACPI fixed event handler pathway, however. It introduced local_irq_disable() workarounds in cmos_check_wkalrm(), which can cause problems on PREEMPT_RT kernels and are now unnecessary.

Add an explicit comment so that this change will not be reverted by mistake.

---
 drivers/rtc/rtc-cmos.c |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/rtc/rtc-cmos.c +6 -4
@@ -692,8 +692,12 @@ static irqreturn_t cmos_interrupt(int ir
 {
 	u8		irqstat;
 	u8		rtc_control;
+	unsigned long	flags;
 
-	spin_lock(&rtc_lock);
+	/* We cannot use spin_lock() here, as cmos_interrupt() is also called
+	 * in a non-irq context.
+	 */
+	spin_lock_irqsave(&rtc_lock, flags);
 
 	/* When the HPET interrupt handler calls us, the interrupt
 	 * status is passed as arg1 instead of the irq number.  But
@@ -727,7 +731,7 @@ static irqreturn_t cmos_interrupt(int ir
 			hpet_mask_rtc_irq_bit(RTC_AIE);
 		CMOS_READ(RTC_INTR_FLAGS);
 	}
-	spin_unlock(&rtc_lock);
+	spin_unlock_irqrestore(&rtc_lock, flags);
 
 	if (is_intr(irqstat)) {
 		rtc_update_irq(p, 1, irqstat);
@@ -1295,9 +1299,7 @@ static void cmos_check_wkalrm(struct dev
 	 * ACK the rtc irq here
 	 */
 	if (t_now >= cmos->alarm_expires && cmos_use_acpi_alarm()) {
-		local_irq_disable();
 		cmos_interrupt(0, (void *)cmos->rtc);
-		local_irq_enable();
 		return;
 	}
 
author maintainer
#5 lore raw

PATCH 5/178 virtio-net: ensure the received length does not exceed allocated size

In reply to Greg Kroah-Hartman

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

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

From: Bui Quang Minh <[email protected]>

commit 315dbdd7cdf6aa533829774caaf4d25f1fd20e73 upstream.

In xdp_linearize_page, when reading the following buffers from the ring, we forget to check the received length with the true allocate size. This can lead to an out-of-bound read. This commit adds that missing check.

---
 drivers/net/virtio_net.c |   38 ++++++++++++++++++++++++++++++++++----
 1 file changed, 34 insertions(+), 4 deletions(-)
drivers/net/virtio_net.c +34 -4
@@ -778,6 +778,26 @@ static unsigned int mergeable_ctx_to_tru
 	return (unsigned long)mrg_ctx & ((1 << MRG_CTX_HEADER_SHIFT) - 1);
 }
 
+static int check_mergeable_len(struct net_device *dev, void *mrg_ctx,
+			       unsigned int len)
+{
+	unsigned int headroom, tailroom, room, truesize;
+
+	truesize = mergeable_ctx_to_truesize(mrg_ctx);
+	headroom = mergeable_ctx_to_headroom(mrg_ctx);
+	tailroom = headroom ? sizeof(struct skb_shared_info) : 0;
+	room = SKB_DATA_ALIGN(headroom + tailroom);
+
+	if (len > truesize - room) {
+		pr_debug("%s: rx error: len %u exceeds truesize %lu\n",
+			 dev->name, len, (unsigned long)(truesize - room));
+		DEV_STATS_INC(dev, rx_length_errors);
+		return -1;
+	}
+
+	return 0;
+}
+
 static struct sk_buff *virtnet_build_skb(void *buf, unsigned int buflen,
 					 unsigned int headroom,
 					 unsigned int len)
@@ -1811,7 +1831,8 @@ static unsigned int virtnet_get_headroom
  * across multiple buffers (num_buf > 1), and we make sure buffers
  * have enough headroom.
  */
-static struct page *xdp_linearize_page(struct receive_queue *rq,
+static struct page *xdp_linearize_page(struct net_device *dev,
+				       struct receive_queue *rq,
 				       int *num_buf,
 				       struct page *p,
 				       int offset,
@@ -1831,18 +1852,27 @@ static struct page *xdp_linearize_page(s
 	memcpy(page_address(page) + page_off, page_address(p) + offset, *len);
 	page_off += *len;
 
+	/* Only mergeable mode can go inside this while loop. In small mode,
+	 * *num_buf == 1, so it cannot go inside.
+	 */
 	while (--*num_buf) {
 		unsigned int buflen;
 		void *buf;
+		void *ctx;
 		int off;
 
-		buf = virtnet_rq_get_buf(rq, &buflen, NULL);
+		buf = virtnet_rq_get_buf(rq, &buflen, &ctx);
 		if (unlikely(!buf))
 			goto err_buf;
 
 		p = virt_to_head_page(buf);
 		off = buf - page_address(p);
 
+		if (check_mergeable_len(dev, ctx, buflen)) {
+			put_page(p);
+			goto err_buf;
+		}
+
 		/* guard against a misconfigured or uncooperative backend that
 		 * is sending packet larger than the MTU.
 		 */
@@ -1931,7 +1961,7 @@ static struct sk_buff *receive_small_xdp
 		headroom = vi->hdr_len + header_offset;
 		buflen = SKB_DATA_ALIGN(GOOD_PACKET_LEN + headroom) +
 			SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
-		xdp_page = xdp_linearize_page(rq, &num_buf, page,
+		xdp_page = xdp_linearize_page(dev, rq, &num_buf, page,
 					      offset, header_offset,
 					      &tlen);
 		if (!xdp_page)
@@ -2266,7 +2296,7 @@ static void *mergeable_xdp_get_buf(struc
 	 */
 	if (!xdp_prog->aux->xdp_has_frags) {
 		/* linearize data for XDP */
-		xdp_page = xdp_linearize_page(rq, num_buf,
+		xdp_page = xdp_linearize_page(vi->dev, rq, num_buf,
 					      *page, offset,
 					      XDP_PACKET_HEADROOM,
 					      len);
author maintainer
#14 lore raw

PATCH 14/178 mmc: sdhci: Add a helper function for dump register in dynamic debug mode

In reply to Greg Kroah-Hartman

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

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

From: Victor Shih <[email protected]>

commit 2881ba9af073faa8ee7408a8d1e0575e50eb3f6c upstream.

Add a helper function for dump register in dynamic debug mode.

---
 drivers/mmc/host/sdhci.h |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/mmc/host/sdhci.h +16 -0
@@ -898,4 +898,20 @@ void sdhci_switch_external_dma(struct sd
 void sdhci_set_data_timeout_irq(struct sdhci_host *host, bool enable);
 void __sdhci_set_timeout(struct sdhci_host *host, struct mmc_command *cmd);
 
+#if defined(CONFIG_DYNAMIC_DEBUG) || \
+	(defined(CONFIG_DYNAMIC_DEBUG_CORE) && defined(DYNAMIC_DEBUG_MODULE))
+#define SDHCI_DBG_ANYWAY 0
+#elif defined(DEBUG)
+#define SDHCI_DBG_ANYWAY 1
+#else
+#define SDHCI_DBG_ANYWAY 0
+#endif
+
+#define sdhci_dbg_dumpregs(host, fmt)					\
+do {									\
+	DEFINE_DYNAMIC_DEBUG_METADATA(descriptor, fmt);			\
+	if (DYNAMIC_DEBUG_BRANCH(descriptor) ||	SDHCI_DBG_ANYWAY)	\
+		sdhci_dumpregs(host);					\
+} while (0)
+
 #endif /* __SDHCI_HW_H */
author maintainer
#15 lore raw

PATCH 15/178 xfs: actually use the xfs_growfs_check_rtgeom tracepoint

In reply to Greg Kroah-Hartman

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

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

From: Darrick J. Wong <[email protected]>

commit db44d088a5ab030b741a3adf2e7b181a8a6dcfbe upstream.

We created a new tracepoint but forgot to put it in. Fix that.

Fixes 59a57acbce282d ("xfs: check that the rtrmapbt maxlevels doesn't increase when growing fs")
---
 fs/xfs/xfs_rtalloc.c | 2 ++
 1 file changed, 2 insertions(+)
fs/xfs/xfs_rtalloc.c +2 -0
index 6484c596ecea..736eb0924573 100644
@@ -1259,6 +1259,8 @@ xfs_growfs_check_rtgeom(
 
 	kfree(nmp);
 
+	trace_xfs_growfs_check_rtgeom(mp, min_logfsbs);
+
 	if (min_logfsbs > mp->m_sb.sb_logblocks)
 		return -EINVAL;
 
Signature · 1 line
2.50.0
author maintainer
#16 lore raw

PATCH 16/178 Revert "mmc: sdhci: Disable SD card clock before changing parameters"

In reply to Greg Kroah-Hartman

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

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

From: Ulf Hansson <[email protected]>

commit dcc3bcfc5b50c625b475dcc25d167b6b947a6637 upstream.

It has turned out the trying to strictly conform to the SDHCI specification is causing problems. Let's revert and start over.

This reverts commit fb3bbc46c94f261b6156ee863c1b06c84cf157dc.

---
 drivers/mmc/host/sdhci.c |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)
drivers/mmc/host/sdhci.c +2 -7
@@ -2065,15 +2065,10 @@ void sdhci_set_clock(struct sdhci_host *
 
 	host->mmc->actual_clock = 0;
 
-	clk = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
-	if (clk & SDHCI_CLOCK_CARD_EN)
-		sdhci_writew(host, clk & ~SDHCI_CLOCK_CARD_EN,
-			SDHCI_CLOCK_CONTROL);
+	sdhci_writew(host, 0, SDHCI_CLOCK_CONTROL);
 
-	if (clock == 0) {
-		sdhci_writew(host, 0, SDHCI_CLOCK_CONTROL);
+	if (clock == 0)
 		return;
-	}
 
 	clk = sdhci_calc_clk(host, clock, &host->mmc->actual_clock);
 	sdhci_enable_clk(host, clk);
author maintainer
#17 lore raw

PATCH 17/178 mmc: sdhci-uhs2: Adjust some error messages and register dump for SD UHS-II card

In reply to Greg Kroah-Hartman

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

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

From: Victor Shih <[email protected]>

commit 49b14db035135341f6cf4de7af6ac2cbc8ad29b6 upstream.

Adjust some error messages to debug mode and register dump to dynamic debug mode to avoid causing misunderstanding it is an error.

---
 drivers/mmc/host/sdhci-uhs2.c |   20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)
drivers/mmc/host/sdhci-uhs2.c +10 -10
@@ -99,8 +99,8 @@ void sdhci_uhs2_reset(struct sdhci_host
 	/* hw clears the bit when it's done */
 	if (read_poll_timeout_atomic(sdhci_readw, val, !(val & mask), 10,
 				     UHS2_RESET_TIMEOUT_100MS, true, host, SDHCI_UHS2_SW_RESET)) {
-		pr_warn("%s: %s: Reset 0x%x never completed. %s: clean reset bit.\n", __func__,
-			mmc_hostname(host->mmc), (int)mask, mmc_hostname(host->mmc));
+		pr_debug("%s: %s: Reset 0x%x never completed. %s: clean reset bit.\n", __func__,
+			 mmc_hostname(host->mmc), (int)mask, mmc_hostname(host->mmc));
 		sdhci_writeb(host, 0, SDHCI_UHS2_SW_RESET);
 		return;
 	}
@@ -335,8 +335,8 @@ static int sdhci_uhs2_interface_detect(s
 	if (read_poll_timeout(sdhci_readl, val, (val & SDHCI_UHS2_IF_DETECT),
 			      100, UHS2_INTERFACE_DETECT_TIMEOUT_100MS, true,
 			      host, SDHCI_PRESENT_STATE)) {
-		pr_warn("%s: not detect UHS2 interface in 100ms.\n", mmc_hostname(host->mmc));
-		sdhci_dumpregs(host);
+		pr_debug("%s: not detect UHS2 interface in 100ms.\n", mmc_hostname(host->mmc));
+		sdhci_dbg_dumpregs(host, "UHS2 interface detect timeout in 100ms");
 		return -EIO;
 	}
 
@@ -345,8 +345,8 @@ static int sdhci_uhs2_interface_detect(s
 
 	if (read_poll_timeout(sdhci_readl, val, (val & SDHCI_UHS2_LANE_SYNC),
 			      100, UHS2_LANE_SYNC_TIMEOUT_150MS, true, host, SDHCI_PRESENT_STATE)) {
-		pr_warn("%s: UHS2 Lane sync fail in 150ms.\n", mmc_hostname(host->mmc));
-		sdhci_dumpregs(host);
+		pr_debug("%s: UHS2 Lane sync fail in 150ms.\n", mmc_hostname(host->mmc));
+		sdhci_dbg_dumpregs(host, "UHS2 Lane sync fail in 150ms");
 		return -EIO;
 	}
 
@@ -417,12 +417,12 @@ static int sdhci_uhs2_do_detect_init(str
 		host->ops->uhs2_pre_detect_init(host);
 
 	if (sdhci_uhs2_interface_detect(host)) {
-		pr_warn("%s: cannot detect UHS2 interface.\n", mmc_hostname(host->mmc));
+		pr_debug("%s: cannot detect UHS2 interface.\n", mmc_hostname(host->mmc));
 		return -EIO;
 	}
 
 	if (sdhci_uhs2_init(host)) {
-		pr_warn("%s: UHS2 init fail.\n", mmc_hostname(host->mmc));
+		pr_debug("%s: UHS2 init fail.\n", mmc_hostname(host->mmc));
 		return -EIO;
 	}
 
@@ -504,8 +504,8 @@ static int sdhci_uhs2_check_dormant(stru
 	if (read_poll_timeout(sdhci_readl, val, (val & SDHCI_UHS2_IN_DORMANT_STATE),
 			      100, UHS2_CHECK_DORMANT_TIMEOUT_100MS, true, host,
 			      SDHCI_PRESENT_STATE)) {
-		pr_warn("%s: UHS2 IN_DORMANT fail in 100ms.\n", mmc_hostname(host->mmc));
-		sdhci_dumpregs(host);
+		pr_debug("%s: UHS2 IN_DORMANT fail in 100ms.\n", mmc_hostname(host->mmc));
+		sdhci_dbg_dumpregs(host, "UHS2 IN_DORMANT fail in 100ms");
 		return -EIO;
 	}
 	return 0;
author maintainer
#18 lore raw

PATCH 18/178 mmc: core: sd: Apply BROKEN_SD_DISCARD quirk earlier

In reply to Greg Kroah-Hartman

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

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

From: Avri Altman <[email protected]>

commit 009c3a4bc41e855fd76f92727f9fbae4e5917d7f upstream.

Move the BROKEN_SD_DISCARD quirk for certain SanDisk SD cards from the mmc_blk_fixups[] to mmc_sd_fixups[]. This ensures the quirk is applied earlier in the device initialization process, aligning with the reasoning in [1]. Applying the quirk sooner prevents the kernel from incorrectly enabling discard support on affected cards during initial setup.

[1] https://lore.kernel.org/all/[email protected]

Fixes 07d2872bf4c8 ("mmc: core: Add SD card quirk for broken discard")
---
 drivers/mmc/core/quirks.h |   12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/mmc/core/quirks.h +6 -6
@@ -44,6 +44,12 @@ static const struct mmc_fixup __maybe_un
 		   0, -1ull, SDIO_ANY_ID, SDIO_ANY_ID, add_quirk_sd,
 		   MMC_QUIRK_NO_UHS_DDR50_TUNING, EXT_CSD_REV_ANY),
 
+	/*
+	 * Some SD cards reports discard support while they don't
+	 */
+	MMC_FIXUP(CID_NAME_ANY, CID_MANFID_SANDISK_SD, 0x5344, add_quirk_sd,
+		  MMC_QUIRK_BROKEN_SD_DISCARD),
+
 	END_FIXUP
 };
 
@@ -147,12 +153,6 @@ static const struct mmc_fixup __maybe_un
 	MMC_FIXUP("M62704", CID_MANFID_KINGSTON, 0x0100, add_quirk_mmc,
 		  MMC_QUIRK_TRIM_BROKEN),
 
-	/*
-	 * Some SD cards reports discard support while they don't
-	 */
-	MMC_FIXUP(CID_NAME_ANY, CID_MANFID_SANDISK_SD, 0x5344, add_quirk_sd,
-		  MMC_QUIRK_BROKEN_SD_DISCARD),
-
 	END_FIXUP
 };
 
author maintainer
#19 lore raw

PATCH 19/178 mmc: core: Adjust some error messages for SD UHS-II cards

In reply to Greg Kroah-Hartman

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

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

From: Victor Shih <[email protected]>

commit 14633da0f416fdbb6844d1b295cdc828b666e273 upstream.

Adjust some error messages to debug mode to avoid causing misunderstanding it is an error.

---
 drivers/mmc/core/sd_uhs2.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/mmc/core/sd_uhs2.c +2 -2
@@ -91,8 +91,8 @@ static int sd_uhs2_phy_init(struct mmc_h
 
 	err = host->ops->uhs2_control(host, UHS2_PHY_INIT);
 	if (err) {
-		pr_err("%s: failed to initial phy for UHS-II!\n",
-		       mmc_hostname(host));
+		pr_debug("%s: failed to initial phy for UHS-II!\n",
+			 mmc_hostname(host));
 	}
 
 	return err;
author maintainer
#20 lore raw

PATCH 20/178 Bluetooth: HCI: Set extended advertising data synchronously

In reply to Greg Kroah-Hartman

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

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

From: Christian Eggers <[email protected]>

commit 89fb8acc38852116d38d721ad394aad7f2871670 upstream.

Currently, for controllers with extended advertising, the advertising data is set in the asynchronous response handler for extended adverstising params. As most advertising settings are performed in a synchronous context, the (asynchronous) setting of the advertising data is done too late (after enabling the advertising).

Move setting of adverstising data from asynchronous response handler into synchronous context to fix ordering of HCI commands.

Fixes a0fb3726ba55 ("Bluetooth: Use Set ext adv/scan rsp data if controller supports")

v2: https://lore.kernel.org/linux-bluetooth/[email protected]/

---
 net/bluetooth/hci_event.c |   36 --------
 net/bluetooth/hci_sync.c  |  207 ++++++++++++++++++++++++++++------------------
 2 files changed, 130 insertions(+), 113 deletions(-)
net/bluetooth/hci_event.c +0 -36
@@ -2150,40 +2150,6 @@ static u8 hci_cc_set_adv_param(struct hc
 	return rp->status;
 }
 
-static u8 hci_cc_set_ext_adv_param(struct hci_dev *hdev, void *data,
-				   struct sk_buff *skb)
-{
-	struct hci_rp_le_set_ext_adv_params *rp = data;
-	struct hci_cp_le_set_ext_adv_params *cp;
-	struct adv_info *adv_instance;
-
-	bt_dev_dbg(hdev, "status 0x%2.2x", rp->status);
-
-	if (rp->status)
-		return rp->status;
-
-	cp = hci_sent_cmd_data(hdev, HCI_OP_LE_SET_EXT_ADV_PARAMS);
-	if (!cp)
-		return rp->status;
-
-	hci_dev_lock(hdev);
-	hdev->adv_addr_type = cp->own_addr_type;
-	if (!cp->handle) {
-		/* Store in hdev for instance 0 */
-		hdev->adv_tx_power = rp->tx_power;
-	} else {
-		adv_instance = hci_find_adv_instance(hdev, cp->handle);
-		if (adv_instance)
-			adv_instance->tx_power = rp->tx_power;
-	}
-	/* Update adv data as tx power is known now */
-	hci_update_adv_data(hdev, cp->handle);
-
-	hci_dev_unlock(hdev);
-
-	return rp->status;
-}
-
 static u8 hci_cc_read_rssi(struct hci_dev *hdev, void *data,
 			   struct sk_buff *skb)
 {
@@ -4164,8 +4130,6 @@ static const struct hci_cc {
 	HCI_CC(HCI_OP_LE_READ_NUM_SUPPORTED_ADV_SETS,
 	       hci_cc_le_read_num_adv_sets,
 	       sizeof(struct hci_rp_le_read_num_supported_adv_sets)),
-	HCI_CC(HCI_OP_LE_SET_EXT_ADV_PARAMS, hci_cc_set_ext_adv_param,
-	       sizeof(struct hci_rp_le_set_ext_adv_params)),
 	HCI_CC_STATUS(HCI_OP_LE_SET_EXT_ADV_ENABLE,
 		      hci_cc_le_set_ext_adv_enable),
 	HCI_CC_STATUS(HCI_OP_LE_SET_ADV_SET_RAND_ADDR,
net/bluetooth/hci_sync.c +130 -77

240 lines not on this page · raw message

author maintainer
#21 lore raw

PATCH 21/178 Bluetooth: hci_sync: revert some mesh modifications

In reply to Greg Kroah-Hartman

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

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

From: Christian Eggers <[email protected]>

commit 46c0d947b64ac8efcf89dd754213dab5d1bd00aa upstream.

This reverts minor parts of the changes made in commit b338d91703fa ("Bluetooth: Implement support for Mesh"). It looks like these changes were only made for development purposes but shouldn't have been part of the commit.

Fixes b338d91703fa ("Bluetooth: Implement support for Mesh")
---
 net/bluetooth/hci_sync.c |   16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)
net/bluetooth/hci_sync.c +4 -12
@@ -2004,13 +2004,10 @@ static int hci_clear_adv_sets_sync(struc
 static int hci_clear_adv_sync(struct hci_dev *hdev, struct sock *sk, bool force)
 {
 	struct adv_info *adv, *n;
-	int err = 0;
 
 	if (ext_adv_capable(hdev))
 		/* Remove all existing sets */
-		err = hci_clear_adv_sets_sync(hdev, sk);
-	if (ext_adv_capable(hdev))
-		return err;
+		return hci_clear_adv_sets_sync(hdev, sk);
 
 	/* This is safe as long as there is no command send while the lock is
 	 * held.
@@ -2038,13 +2035,11 @@ static int hci_clear_adv_sync(struct hci
 static int hci_remove_adv_sync(struct hci_dev *hdev, u8 instance,
 			       struct sock *sk)
 {
-	int err = 0;
+	int err;
 
 	/* If we use extended advertising, instance has to be removed first. */
 	if (ext_adv_capable(hdev))
-		err = hci_remove_ext_adv_instance_sync(hdev, instance, sk);
-	if (ext_adv_capable(hdev))
-		return err;
+		return hci_remove_ext_adv_instance_sync(hdev, instance, sk);
 
 	/* This is safe as long as there is no command send while the lock is
 	 * held.
@@ -2143,16 +2138,13 @@ int hci_read_tx_power_sync(struct hci_de
 int hci_disable_advertising_sync(struct hci_dev *hdev)
 {
 	u8 enable = 0x00;
-	int err = 0;
 
 	/* If controller is not advertising we are done. */
 	if (!hci_dev_test_flag(hdev, HCI_LE_ADV))
 		return 0;
 
 	if (ext_adv_capable(hdev))
-		err = hci_disable_ext_adv_instance_sync(hdev, 0x00);
-	if (ext_adv_capable(hdev))
-		return err;
+		return hci_disable_ext_adv_instance_sync(hdev, 0x00);
 
 	return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_ADV_ENABLE,
 				     sizeof(enable), &enable, HCI_CMD_TIMEOUT);
author maintainer
#22 lore raw

PATCH 22/178 Bluetooth: MGMT: set_mesh: update LE scan interval and window

In reply to Greg Kroah-Hartman

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

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

From: Christian Eggers <[email protected]>

commit e5af67a870f738bb8a4594b6c60c2caf4c87a3c9 upstream.

According to the message of commit b338d91703fa ("Bluetooth: Implement support for Mesh"), MGMT_OP_SET_MESH_RECEIVER should set the passive scan parameters. Currently the scan interval and window parameters are silently ignored, although user space (bluetooth-meshd) expects that they can be used [1]

[1] https://git.kernel.org/pub/scm/bluetooth/bluez.git/tree/mesh/mesh-io-mgmt.c#n344

Fixes b338d91703fa ("Bluetooth: Implement support for Mesh")
---
 net/bluetooth/mgmt.c |   22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)
net/bluetooth/mgmt.c +22 -0
@@ -2153,6 +2153,9 @@ static int set_mesh_sync(struct hci_dev
 	else
 		hci_dev_clear_flag(hdev, HCI_MESH);
 
+	hdev->le_scan_interval = __le16_to_cpu(cp->period);
+	hdev->le_scan_window = __le16_to_cpu(cp->window);
+
 	len -= sizeof(*cp);
 
 	/* If filters don't fit, forward all adv pkts */
@@ -2167,6 +2170,7 @@ static int set_mesh(struct sock *sk, str
 {
 	struct mgmt_cp_set_mesh *cp = data;
 	struct mgmt_pending_cmd *cmd;
+	__u16 period, window;
 	int err = 0;
 
 	bt_dev_dbg(hdev, "sock %p", sk);
@@ -2180,6 +2184,23 @@ static int set_mesh(struct sock *sk, str
 		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_MESH_RECEIVER,
 				       MGMT_STATUS_INVALID_PARAMS);
 
+	/* Keep allowed ranges in sync with set_scan_params() */
+	period = __le16_to_cpu(cp->period);
+
+	if (period < 0x0004 || period > 0x4000)
+		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_MESH_RECEIVER,
+				       MGMT_STATUS_INVALID_PARAMS);
+
+	window = __le16_to_cpu(cp->window);
+
+	if (window < 0x0004 || window > 0x4000)
+		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_MESH_RECEIVER,
+				       MGMT_STATUS_INVALID_PARAMS);
+
+	if (window > period)
+		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_MESH_RECEIVER,
+				       MGMT_STATUS_INVALID_PARAMS);
+
 	hci_dev_lock(hdev);
 
 	cmd = mgmt_pending_add(sk, MGMT_OP_SET_MESH_RECEIVER, hdev, data, len);
@@ -6432,6 +6453,7 @@ static int set_scan_params(struct sock *
 		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_SCAN_PARAMS,
 				       MGMT_STATUS_NOT_SUPPORTED);
 
+	/* Keep allowed ranges in sync with set_mesh() */
 	interval = __le16_to_cpu(cp->interval);
 
 	if (interval < 0x0004 || interval > 0x4000)
author maintainer
#23 lore raw

PATCH 23/178 Bluetooth: MGMT: mesh_send: check instances prior disabling advertising

In reply to Greg Kroah-Hartman

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

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

From: Christian Eggers <[email protected]>

commit f3cb5676e5c11c896ba647ee309a993e73531588 upstream.

The unconditional call of hci_disable_advertising_sync() in mesh_send_done_sync() also disables other LE advertisings (non mesh related).

I am not sure whether this call is required at all, but checking the adv_instances list (like done at other places) seems to solve the problem.

Fixes b338d91703fa ("Bluetooth: Implement support for Mesh")
---
 net/bluetooth/mgmt.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/bluetooth/mgmt.c +2 -1
@@ -1080,7 +1080,8 @@ static int mesh_send_done_sync(struct hc
 	struct mgmt_mesh_tx *mesh_tx;
 
 	hci_dev_clear_flag(hdev, HCI_MESH_SENDING);
-	hci_disable_advertising_sync(hdev);
+	if (list_empty(&hdev->adv_instances))
+		hci_disable_advertising_sync(hdev);
 	mesh_tx = mgmt_mesh_next(hdev, NULL);
 
 	if (mesh_tx)
author maintainer
#6 lore raw

PATCH 6/178 s390/pci: Fix stale function handles in error handling

In reply to Greg Kroah-Hartman

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

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

From: Niklas Schnelle <[email protected]>

commit 45537926dd2aaa9190ac0fac5a0fbeefcadfea95 upstream.

The error event information for PCI error events contains a function handle for the respective function. This handle is generally captured at the time the error event was recorded. Due to delays in processing or cascading issues, it may happen that during firmware recovery multiple events are generated. When processing these events in order Linux may already have recovered an affected function making the event information stale. Fix this by doing an unconditional CLP List PCI function retrieving the current function handle with the zdev->state_lock held and ignoring the event if its function handle is stale.

---
 arch/s390/pci/pci_event.c |   11 +++++++++++
 1 file changed, 11 insertions(+)
arch/s390/pci/pci_event.c +11 -0
@@ -273,6 +273,8 @@ static void __zpci_event_error(struct zp
 	struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid);
 	struct pci_dev *pdev = NULL;
 	pci_ers_result_t ers_res;
+	u32 fh = 0;
+	int rc;
 
 	zpci_dbg(3, "err fid:%x, fh:%x, pec:%x\n",
 		 ccdf->fid, ccdf->fh, ccdf->pec);
@@ -281,6 +283,15 @@ static void __zpci_event_error(struct zp
 
 	if (zdev) {
 		mutex_lock(&zdev->state_lock);
+		rc = clp_refresh_fh(zdev->fid, &fh);
+		if (rc)
+			goto no_pdev;
+		if (!fh || ccdf->fh != fh) {
+			/* Ignore events with stale handles */
+			zpci_dbg(3, "err fid:%x, fh:%x (stale %x)\n",
+				 ccdf->fid, fh, ccdf->fh);
+			goto no_pdev;
+		}
 		zpci_update_fh(zdev, ccdf->fh);
 		if (zdev->zbus->bus)
 			pdev = pci_get_slot(zdev->zbus->bus, zdev->devfn);
author maintainer
#24 lore raw

PATCH 24/178 iommufd/selftest: Fix iommufd_dirty_tracking with large hugepage sizes

In reply to Greg Kroah-Hartman

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

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

From: Nicolin Chen <[email protected]>

commit 818625570558cd91082c9bafd6f2b59b73241a69 upstream.

The hugepage test cases of iommufd_dirty_tracking have the 64MB and 128MB coverages. Both of them are smaller than the default hugepage size 512MB, when CONFIG_PAGE_SIZE_64KB=y. However, these test cases have a variant of using huge pages, which would mmap(MAP_HUGETLB) using these smaller sizes than the system hugepag size. This results in the kernel aligning up the smaller size to 512MB. If a memory was located between the upper 64/128MB size boundary and the hugepage 512MB boundary, it would get wiped out:
https://lore.kernel.org/all/[email protected]/

Given that this aligning up behavior is well documented, we have no choice but to allocate a hugepage aligned size to avoid this unintended wipe out. Instead of relying on the kernel's internal force alignment, pass the same size to posix_memalign() and map().

Also, fix the FIXTURE_TEARDOWN() misusing munmap() to free the memory from posix_memalign(), as munmap() doesn't destroy the allocator meta data. So, call free() instead.

---
 tools/testing/selftests/iommu/iommufd.c |   30 +++++++++++++++++++++---------
 1 file changed, 21 insertions(+), 9 deletions(-)
tools/testing/selftests/iommu/iommufd.c +21 -9
@@ -2010,6 +2010,7 @@ FIXTURE_VARIANT(iommufd_dirty_tracking)
 
 FIXTURE_SETUP(iommufd_dirty_tracking)
 {
+	size_t mmap_buffer_size;
 	unsigned long size;
 	int mmap_flags;
 	void *vrc;
@@ -2024,22 +2025,33 @@ FIXTURE_SETUP(iommufd_dirty_tracking)
 	self->fd = open("/dev/iommu", O_RDWR);
 	ASSERT_NE(-1, self->fd);
 
-	rc = posix_memalign(&self->buffer, HUGEPAGE_SIZE, variant->buffer_size);
-	if (rc || !self->buffer) {
-		SKIP(return, "Skipping buffer_size=%lu due to errno=%d",
-			   variant->buffer_size, rc);
-	}
-
 	mmap_flags = MAP_SHARED | MAP_ANONYMOUS | MAP_FIXED;
+	mmap_buffer_size = variant->buffer_size;
 	if (variant->hugepages) {
 		/*
 		 * MAP_POPULATE will cause the kernel to fail mmap if THPs are
 		 * not available.
 		 */
 		mmap_flags |= MAP_HUGETLB | MAP_POPULATE;
+
+		/*
+		 * Allocation must be aligned to the HUGEPAGE_SIZE, because the
+		 * following mmap() will automatically align the length to be a
+		 * multiple of the underlying huge page size. Failing to do the
+		 * same at this allocation will result in a memory overwrite by
+		 * the mmap().
+		 */
+		if (mmap_buffer_size < HUGEPAGE_SIZE)
+			mmap_buffer_size = HUGEPAGE_SIZE;
+	}
+
+	rc = posix_memalign(&self->buffer, HUGEPAGE_SIZE, mmap_buffer_size);
+	if (rc || !self->buffer) {
+		SKIP(return, "Skipping buffer_size=%lu due to errno=%d",
+			   mmap_buffer_size, rc);
 	}
 	assert((uintptr_t)self->buffer % HUGEPAGE_SIZE == 0);
-	vrc = mmap(self->buffer, variant->buffer_size, PROT_READ | PROT_WRITE,
+	vrc = mmap(self->buffer, mmap_buffer_size, PROT_READ | PROT_WRITE,
 		   mmap_flags, -1, 0);
 	assert(vrc == self->buffer);
 
@@ -2068,8 +2080,8 @@ FIXTURE_SETUP(iommufd_dirty_tracking)
 
 FIXTURE_TEARDOWN(iommufd_dirty_tracking)
 {
-	munmap(self->buffer, variant->buffer_size);
-	munmap(self->bitmap, DIV_ROUND_UP(self->bitmap_size, BITS_PER_BYTE));
+	free(self->buffer);
+	free(self->bitmap);
 	teardown_iommufd(self->fd, _metadata);
 }
 
author maintainer
#25 lore raw

PATCH 25/178 regulator: gpio: Fix the out-of-bounds access to drvdata::gpiods

In reply to Greg Kroah-Hartman

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

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

From: Manivannan Sadhasivam <[email protected]>

commit c9764fd88bc744592b0604ccb6b6fc1a5f76b4e3 upstream.

drvdata::gpiods is supposed to hold an array of 'gpio_desc' pointers. But the memory is allocated for only one pointer. This will lead to out-of-bounds access later in the code if 'config::ngpios' is > 1. So fix the code to allocate enough memory to hold 'config::ngpios' of GPIO descriptors.

While at it, also move the check for memory allocation failure to be below the allocation to make it more readable.

Fixes d6cd33ad7102 ("regulator: gpio: Convert to use descriptors")
---
 drivers/regulator/gpio-regulator.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/regulator/gpio-regulator.c +4 -4
@@ -260,8 +260,10 @@ static int gpio_regulator_probe(struct p
 		return -ENOMEM;
 	}
 
-	drvdata->gpiods = devm_kzalloc(dev, sizeof(struct gpio_desc *),
-				       GFP_KERNEL);
+	drvdata->gpiods = devm_kcalloc(dev, config->ngpios,
+				       sizeof(struct gpio_desc *), GFP_KERNEL);
+	if (!drvdata->gpiods)
+		return -ENOMEM;
 
 	if (config->input_supply) {
 		drvdata->desc.supply_name = devm_kstrdup(&pdev->dev,
@@ -274,8 +276,6 @@ static int gpio_regulator_probe(struct p
 		}
 	}
 
-	if (!drvdata->gpiods)
-		return -ENOMEM;
 	for (i = 0; i < config->ngpios; i++) {
 		drvdata->gpiods[i] = devm_gpiod_get_index(dev,
 							  NULL,
author maintainer
#26 lore raw

PATCH 26/178 Input: cs40l50-vibra - fix potential NULL dereference in cs40l50_upload_owt()

In reply to Greg Kroah-Hartman

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

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

From: Yunshui Jiang <[email protected]>

commit 4cf65845fdd09d711fc7546d60c9abe010956922 upstream.

The cs40l50_upload_owt() function allocates memory via kmalloc() without checking for allocation failure, which could lead to a NULL pointer dereference.

Return -ENOMEM in case allocation fails.

Fixes c38fe1bb5d21 ("Input: cs40l50 - Add support for the CS40L50 haptic driver")
---
 drivers/input/misc/cs40l50-vibra.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/input/misc/cs40l50-vibra.c +2 -0
@@ -238,6 +238,8 @@ static int cs40l50_upload_owt(struct cs4
 	header.data_words = len / sizeof(u32);
 
 	new_owt_effect_data = kmalloc(sizeof(header) + len, GFP_KERNEL);
+	if (!new_owt_effect_data)
+		return -ENOMEM;
 
 	memcpy(new_owt_effect_data, &header, sizeof(header));
 	memcpy(new_owt_effect_data + sizeof(header), work_data->custom_data, len);
author maintainer
#27 lore raw

PATCH 27/178 anon_inode: rework assertions

In reply to Greg Kroah-Hartman

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

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

From: Christian Brauner <[email protected]>

commit 1e7ab6f67824343ee3e96f100f0937c393749a8a upstream.

Making anonymous inodes regular files comes with a lot of risk and regression potential as evidenced by a recent hickup in io_uring. We're better of continuing to not have them be regular files. Since we have S_ANON_INODE we can port all of our assertions easily.

---
 fs/exec.c  |    9 +++++++--
 fs/libfs.c |    8 +++-----
 fs/namei.c |    2 +-
 3 files changed, 11 insertions(+), 8 deletions(-)
fs/exec.c +7 -2
@@ -111,6 +111,9 @@ static inline void put_binfmt(struct lin
 
 bool path_noexec(const struct path *path)
 {
+	/* If it's an anonymous inode make sure that we catch any shenanigans. */
+	VFS_WARN_ON_ONCE(IS_ANON_FILE(d_inode(path->dentry)) &&
+			 !(path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC));
 	return (path->mnt->mnt_flags & MNT_NOEXEC) ||
 	       (path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC);
 }
@@ -894,13 +897,15 @@ static struct file *do_open_execat(int f
 	if (IS_ERR(file))
 		return file;
 
+	if (path_noexec(&file->f_path))
+		return ERR_PTR(-EACCES);
+
 	/*
 	 * In the past the regular type check was here. It moved to may_open() in
 	 * 633fb6ac3980 ("exec: move S_ISREG() check earlier"). Since then it is
 	 * an invariant that all non-regular files error out before we get here.
 	 */
-	if (WARN_ON_ONCE(!S_ISREG(file_inode(file)->i_mode)) ||
-	    path_noexec(&file->f_path))
+	if (WARN_ON_ONCE(!S_ISREG(file_inode(file)->i_mode)))
 		return ERR_PTR(-EACCES);
 
 	err = exe_file_deny_write_access(file);
fs/libfs.c +3 -5
@@ -1648,12 +1648,10 @@ struct inode *alloc_anon_inode(struct su
 	 */
 	inode->i_state = I_DIRTY;
 	/*
-	 * Historically anonymous inodes didn't have a type at all and
-	 * userspace has come to rely on this. Internally they're just
-	 * regular files but S_IFREG is masked off when reporting
-	 * information to userspace.
+	 * Historically anonymous inodes don't have a type at all and
+	 * userspace has come to rely on this.
 	 */
-	inode->i_mode = S_IFREG | S_IRUSR | S_IWUSR;
+	inode->i_mode = S_IRUSR | S_IWUSR;
 	inode->i_uid = current_fsuid();
 	inode->i_gid = current_fsgid();
 	inode->i_flags |= S_PRIVATE | S_ANON_INODE;
fs/namei.c +1 -1
@@ -3464,7 +3464,7 @@ static int may_open(struct mnt_idmap *id
 			return -EACCES;
 		break;
 	default:
-		VFS_BUG_ON_INODE(1, inode);
+		VFS_BUG_ON_INODE(!IS_ANON_FILE(inode), inode);
 	}
 
 	error = inode_permission(idmap, inode, MAY_OPEN | acc_mode);
author maintainer
#28 lore raw

PATCH 28/178 usb: typec: altmodes/displayport: do not index invalid pin_assignments

In reply to Greg Kroah-Hartman

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

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

From: RD Babiera <[email protected]>

commit af4db5a35a4ef7a68046883bfd12468007db38f1 upstream.

A poorly implemented DisplayPort Alt Mode port partner can indicate that its pin assignment capabilities are greater than the maximum value, DP_PIN_ASSIGN_F. In this case, calls to pin_assignment_show will cause a BRK exception due to an out of bounds array access.

Prevent for loop in pin_assignment_show from accessing invalid values in pin_assignments by adding DP_PIN_ASSIGN_MAX value in typec_dp.h and using i < DP_PIN_ASSIGN_MAX as a loop condition.

---
 drivers/usb/typec/altmodes/displayport.c |    2 +-
 include/linux/usb/typec_dp.h             |    1 +
 2 files changed, 2 insertions(+), 1 deletion(-)
drivers/usb/typec/altmodes/displayport.c +1 -1
@@ -677,7 +677,7 @@ static ssize_t pin_assignment_show(struc
 
 	assignments = get_current_pin_assignments(dp);
 
-	for (i = 0; assignments; assignments >>= 1, i++) {
+	for (i = 0; assignments && i < DP_PIN_ASSIGN_MAX; assignments >>= 1, i++) {
 		if (assignments & 1) {
 			if (i == cur)
 				len += sprintf(buf + len, "[%s] ",
include/linux/usb/typec_dp.h +1 -0
@@ -57,6 +57,7 @@ enum {
 	DP_PIN_ASSIGN_D,
 	DP_PIN_ASSIGN_E,
 	DP_PIN_ASSIGN_F, /* Not supported after v1.0b */
+	DP_PIN_ASSIGN_MAX,
 };
 
 /* DisplayPort alt mode specific commands */
author maintainer
#29 lore raw

PATCH 29/178 mtk-sd: Fix a pagefault in dma_unmap_sg() for not prepared data

In reply to Greg Kroah-Hartman

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

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

From: Masami Hiramatsu (Google) <[email protected]>

commit 539d80575b810c7a5987c7ac8915e3bc99c03695 upstream.

When swiotlb buffer is full, the dma_map_sg() returns 0 to msdc_prepare_data(), but it does not check it and sets the MSDC_PREPARE_FLAG.

swiotlb_tbl_map_single() /* prints "swiotlb buffer is full" */
<-swiotlb_map()

    <-dma_direct_map_page()
      <-dma_direct_map_sg()
        <-__dma_map_sg_attrs()
          <-dma_map_sg_attrs()
            <-dma_map_sg()  /* returns 0 (pages mapped) */
              <-msdc_prepare_data()

Then, the msdc_unprepare_data() checks MSDC_PREPARE_FLAG and calls dma_unmap_sg() with unmapped pages. It causes a page fault.

To fix this problem, Do not set MSDC_PREPARE_FLAG if dma_map_sg() fails because this is not prepared.

---
 drivers/mmc/host/mtk-sd.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/mmc/host/mtk-sd.c +2 -1
@@ -827,9 +827,10 @@ static inline void msdc_dma_setup(struct
 static void msdc_prepare_data(struct msdc_host *host, struct mmc_data *data)
 {
 	if (!(data->host_cookie & MSDC_PREPARE_FLAG)) {
-		data->host_cookie |= MSDC_PREPARE_FLAG;
 		data->sg_count = dma_map_sg(host->dev, data->sg, data->sg_len,
 					    mmc_get_dma_dir(data));
+		if (data->sg_count)
+			data->host_cookie |= MSDC_PREPARE_FLAG;
 	}
 }
 
author maintainer
#30 lore raw

PATCH 30/178 mtk-sd: Prevent memory corruption from DMA map failure

In reply to Greg Kroah-Hartman

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

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

From: Masami Hiramatsu (Google) <[email protected]>

commit f5de469990f19569627ea0dd56536ff5a13beaa3 upstream.

If msdc_prepare_data() fails to map the DMA region, the request is not prepared for data receiving, but msdc_start_data() proceeds the DMA with previous setting.
Since this will lead a memory corruption, we have to stop the request operation soon after the msdc_prepare_data() fails to prepare it.

---
 drivers/mmc/host/mtk-sd.c |   17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)
drivers/mmc/host/mtk-sd.c +16 -1
@@ -834,6 +834,11 @@ static void msdc_prepare_data(struct msd
 	}
 }
 
+static bool msdc_data_prepared(struct mmc_data *data)
+{
+	return data->host_cookie & MSDC_PREPARE_FLAG;
+}
+
 static void msdc_unprepare_data(struct msdc_host *host, struct mmc_data *data)
 {
 	if (data->host_cookie & MSDC_ASYNC_FLAG)
@@ -1466,8 +1471,18 @@ static void msdc_ops_request(struct mmc_
 	WARN_ON(!host->hsq_en && host->mrq);
 	host->mrq = mrq;
 
-	if (mrq->data)
+	if (mrq->data) {
 		msdc_prepare_data(host, mrq->data);
+		if (!msdc_data_prepared(mrq->data)) {
+			/*
+			 * Failed to prepare DMA area, fail fast before
+			 * starting any commands.
+			 */
+			mrq->cmd->error = -ENOSPC;
+			mmc_request_done(mmc_from_priv(host), mrq);
+			return;
+		}
+	}
 
 	/* if SBC is required, we have HW option and SW option.
 	 * if HW option is enabled, and SBC does not have "special" flags,
author maintainer
#31 lore raw

PATCH 31/178 mtk-sd: reset host->mrq on prepare_data() error

In reply to Greg Kroah-Hartman

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

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

From: Sergey Senozhatsky <[email protected]>

commit ec54c0a20709ed6e56f40a8d59eee725c31a916b upstream.

Do not leave host with dangling ->mrq pointer if we hit the msdc_prepare_data() error out path.

---
 drivers/mmc/host/mtk-sd.c |    1 +
 1 file changed, 1 insertion(+)
drivers/mmc/host/mtk-sd.c +1 -0
@@ -1474,6 +1474,7 @@ static void msdc_ops_request(struct mmc_
 	if (mrq->data) {
 		msdc_prepare_data(host, mrq->data);
 		if (!msdc_data_prepared(mrq->data)) {
+			host->mrq = NULL;
 			/*
 			 * Failed to prepare DMA area, fail fast before
 			 * starting any commands.
author maintainer
#32 lore raw

PATCH 32/178 drm/v3d: Disable interrupts before resetting the GPU

In reply to Greg Kroah-Hartman

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

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

From: Maíra Canal <[email protected]>

commit 226862f50a7a88e4e4de9abbf36c64d19acd6fd0 upstream.

Currently, an interrupt can be triggered during a GPU reset, which can lead to GPU hangs and NULL pointer dereference in an interrupt context as shown in the following trace:

 [  314.035040] Unable to handle kernel NULL pointer dereference at virtual address 00000000000000c0
 [  314.043822] Mem abort info:
 [  314.046606]   ESR = 0x0000000096000005
 [  314.050347]   EC = 0x25: DABT (current EL), IL = 32 bits
 [  314.055651]   SET = 0, FnV = 0
 [  314.058695]   EA = 0, S1PTW = 0
 [  314.061826]   FSC = 0x05: level 1 translation fault
 [  314.066694] Data abort info:
 [  314.069564]   ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000
 [  314.075039]   CM = 0, WnR = 0, TnD = 0, TagAccess = 0
 [  314.080080]   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
 [  314.085382] user pgtable: 4k pages, 39-bit VAs, pgdp=0000000102728000
 [  314.091814] [00000000000000c0] pgd=0000000000000000, p4d=0000000000000000, pud=0000000000000000
 [  314.100511] Internal error: Oops: 0000000096000005 [#1] PREEMPT SMP
 [  314.106770] Modules linked in: v3d i2c_brcmstb vc4 snd_soc_hdmi_codec gpu_sched drm_shmem_helper drm_display_helper cec drm_dma_helper drm_kms_helper drm drm_panel_orientation_quirks snd_soc_core snd_compress snd_pcm_dmaengine snd_pcm snd_timer snd backlight
 [  314.129654] CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 6.12.25+rpt-rpi-v8 #1  Debian 1:6.12.25-1+rpt1
 [  314.139388] Hardware name: Raspberry Pi 4 Model B Rev 1.4 (DT)
 [  314.145211] pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
 [  314.152165] pc : v3d_irq+0xec/0x2e0 [v3d]
 [  314.156187] lr : v3d_irq+0xe0/0x2e0 [v3d]
 [  314.160198] sp : ffffffc080003ea0
 [  314.163502] x29: ffffffc080003ea0 x28: ffffffec1f184980 x27: 021202b000000000
 [  314.170633] x26: ffffffec1f17f630 x25: ffffff8101372000 x24: ffffffec1f17d9f0
 [  314.177764] x23: 000000000000002a x22: 000000000000002a x21: ffffff8103252000
 [  314.184895] x20: 0000000000000001 x19: 00000000deadbeef x18: 0000000000000000
 [  314.192026] x17: ffffff94e51d2000 x16: ffffffec1dac3cb0 x15: c306000000000000
 [  314.199156] x14: 0000000000000000 x13: b2fc982e03cc5168 x12: 0000000000000001
 [  314.206286] x11: ffffff8103f8bcc0 x10: ffffffec1f196868 x9 : ffffffec1dac3874
 [  314.213416] x8 : 0000000000000000 x7 : 0000000000042a3a x6 : ffffff810017a180
 [  314.220547] x5 : ffffffec1ebad400 x4 : ffffffec1ebad320 x3 : 00000000000bebeb
 [  314.227677] x2 : 0000000000000000 x1 : 0000000000000000 x0 : 0000000000000000
 [  314.234807] Call trace:
 [  314.237243]  v3d_irq+0xec/0x2e0 [v3d]
 [  314.240906]  __handle_irq_event_percpu+0x58/0x218
 [  314.245609]  handle_irq_event+0x54/0xb8
 [  314.249439]  handle_fasteoi_irq+0xac/0x240
 [  314.253527]  handle_irq_desc+0x48/0x68
 [  314.257269]  generic_handle_domain_irq+0x24/0x38
 [  314.261879]  gic_handle_irq+0x48/0xd8
 [  314.265533]  call_on_irq_stack+0x24/0x58
 [  314.269448]  do_interrupt_handler+0x88/0x98
 [  314.273624]  el1_interrupt+0x34/0x68
 [  314.277193]  el1h_64_irq_handler+0x18/0x28
 [  314.281281]  el1h_64_irq+0x64/0x68
 [  314.284673]  default_idle_call+0x3c/0x168
 [  314.288675]  do_idle+0x1fc/0x230
 [  314.291895]  cpu_startup_entry+0x3c/0x50
 [  314.295810]  rest_init+0xe4/0xf0
 [  314.299030]  start_kernel+0x5e8/0x790
 [  314.302684]  __primary_switched+0x80/0x90
 [  314.306691] Code: 940029eb 360ffc13 f9442ea0 52800001 (f9406017)
 [  314.312775] ---[ end trace 0000000000000000 ]---
 [  314.317384] Kernel panic - not syncing: Oops: Fatal exception in interrupt
 [  314.324249] SMP: stopping secondary CPUs
 [  314.328167] Kernel Offset: 0x2b9da00000 from 0xffffffc080000000
 [  314.334076] PHYS_OFFSET: 0x0
 [  314.336946] CPU features: 0x08,00002013,c0200000,0200421b
 [  314.342337] Memory Limit: none
 [  314.345382] ---[ end Kernel panic - not syncing: Oops: Fatal exception in interrupt ]---

Before resetting the GPU, it's necessary to disable all interrupts and deal with any interrupt handler still in-flight. Otherwise, the GPU might reset with jobs still running, or yet, an interrupt could be handled during the reset.

---
 drivers/gpu/drm/v3d/v3d_drv.h |    8 ++++++++
 drivers/gpu/drm/v3d/v3d_gem.c |    2 ++
 drivers/gpu/drm/v3d/v3d_irq.c |   37 +++++++++++++++++++++++++++----------
 3 files changed, 37 insertions(+), 10 deletions(-)
drivers/gpu/drm/v3d/v3d_drv.h +8 -0
@@ -101,6 +101,12 @@ enum v3d_gen {
 	V3D_GEN_71 = 71,
 };
 
+enum v3d_irq {
+	V3D_CORE_IRQ,
+	V3D_HUB_IRQ,
+	V3D_MAX_IRQS,
+};
+
 struct v3d_dev {
 	struct drm_device drm;
 
@@ -112,6 +118,8 @@ struct v3d_dev {
 
 	bool single_irq_line;
 
+	int irq[V3D_MAX_IRQS];
+
 	struct v3d_perfmon_info perfmon_info;
 
 	void __iomem *hub_regs;
drivers/gpu/drm/v3d/v3d_gem.c +2 -0
@@ -118,6 +118,8 @@ v3d_reset(struct v3d_dev *v3d)
 	if (false)
 		v3d_idle_axi(v3d, 0);
 
+	v3d_irq_disable(v3d);
+
 	v3d_idle_gca(v3d);
 	v3d_reset_v3d(v3d);
 
drivers/gpu/drm/v3d/v3d_irq.c +27 -10
@@ -260,7 +260,7 @@ v3d_hub_irq(int irq, void *arg)
 int
 v3d_irq_init(struct v3d_dev *v3d)
 {
-	int irq1, ret, core;
+	int irq, ret, core;
 
 	INIT_WORK(&v3d->overflow_mem_work, v3d_overflow_mem_work);
 
@@ -271,17 +271,24 @@ v3d_irq_init(struct v3d_dev *v3d)
 		V3D_CORE_WRITE(core, V3D_CTL_INT_CLR, V3D_CORE_IRQS(v3d->ver));
 	V3D_WRITE(V3D_HUB_INT_CLR, V3D_HUB_IRQS(v3d->ver));
 
-	irq1 = platform_get_irq_optional(v3d_to_pdev(v3d), 1);
-	if (irq1 == -EPROBE_DEFER)
-		return irq1;
-	if (irq1 > 0) {
-		ret = devm_request_irq(v3d->drm.dev, irq1,
+	irq = platform_get_irq_optional(v3d_to_pdev(v3d), 1);
+	if (irq == -EPROBE_DEFER)
+		return irq;
+	if (irq > 0) {
+		v3d->irq[V3D_CORE_IRQ] = irq;
+
+		ret = devm_request_irq(v3d->drm.dev, v3d->irq[V3D_CORE_IRQ],
 				       v3d_irq, IRQF_SHARED,
 				       "v3d_core0", v3d);
 		if (ret)
 			goto fail;
-		ret = devm_request_irq(v3d->drm.dev,
-				       platform_get_irq(v3d_to_pdev(v3d), 0),
+
+		irq = platform_get_irq(v3d_to_pdev(v3d), 0);
+		if (irq < 0)
+			return irq;
+		v3d->irq[V3D_HUB_IRQ] = irq;
+
+		ret = devm_request_irq(v3d->drm.dev, v3d->irq[V3D_HUB_IRQ],
 				       v3d_hub_irq, IRQF_SHARED,
 				       "v3d_hub", v3d);
 		if (ret)
@@ -289,8 +296,12 @@ v3d_irq_init(struct v3d_dev *v3d)
 	} else {
 		v3d->single_irq_line = true;
 
-		ret = devm_request_irq(v3d->drm.dev,
-				       platform_get_irq(v3d_to_pdev(v3d), 0),
+		irq = platform_get_irq(v3d_to_pdev(v3d), 0);
+		if (irq < 0)
+			return irq;
+		v3d->irq[V3D_CORE_IRQ] = irq;
+
+		ret = devm_request_irq(v3d->drm.dev, v3d->irq[V3D_CORE_IRQ],
 				       v3d_irq, IRQF_SHARED,
 				       "v3d", v3d);
 		if (ret)
@@ -331,6 +342,12 @@ v3d_irq_disable(struct v3d_dev *v3d)
 		V3D_CORE_WRITE(core, V3D_CTL_INT_MSK_SET, ~0);
 	V3D_WRITE(V3D_HUB_INT_MSK_SET, ~0);
 
+	/* Finish any interrupt handler still in flight. */
+	for (int i = 0; i < V3D_MAX_IRQS; i++) {
+		if (v3d->irq[i])
+			synchronize_irq(v3d->irq[i]);
+	}
+
 	/* Clear any pending interrupts we might have left. */
 	for (core = 0; core < v3d->cores; core++)
 		V3D_CORE_WRITE(core, V3D_CTL_INT_CLR, V3D_CORE_IRQS(v3d->ver));
author maintainer
#33 lore raw

PATCH 33/178 firmware: arm_ffa: Fix memory leak by freeing notifier callback node

In reply to Greg Kroah-Hartman

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

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

From: Sudeep Holla <[email protected]>

[ Upstream commit a833d31ad867103ba72a0b73f3606f4ab8601719 ]

Commit e0573444edbf ("firmware: arm_ffa: Add interfaces to request notification callbacks") adds support for notifier callbacks by allocating and inserting a callback node into a hashtable during registration of notifiers. However, during unregistration, the code only removes the node from the hashtable without freeing the associated memory, resulting in a memory leak.

Resolve the memory leak issue by ensuring the allocated notifier callback node is properly freed after it is removed from the hashtable entry.

Fixes e0573444edbf ("firmware: arm_ffa: Add interfaces to request notification callbacks")

Message-Id: <[email protected]>

---
 drivers/firmware/arm_ffa/driver.c | 1 +
 1 file changed, 1 insertion(+)
drivers/firmware/arm_ffa/driver.c +1 -0
index fe55613a8ea99..6f75cdf297209 100644
@@ -1284,6 +1284,7 @@ update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb,
 		hash_add(drv_info->notifier_hash, &cb_info->hnode, notify_id);
 	} else {
 		hash_del(&cb_info->hnode);
+		kfree(cb_info);
 	}
 
 	return 0;
Signature · 1 line
2.39.5
author maintainer
#7 lore raw

PATCH 7/178 s390/pci: Do not try re-enabling load/store if device is disabled

In reply to Greg Kroah-Hartman

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

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

From: Niklas Schnelle <[email protected]>

commit b97a7972b1f4f81417840b9a2ab0c19722b577d5 upstream.

If a device is disabled unblocking load/store on its own is not useful as a full re-enable of the function is necessary anyway. Note that SCLP Write Event Data Action Qualifier 0 (Reset) leaves the device disabled and triggers this case unless the driver already requests a reset.

Fixes 4cdf2f4e24ff ("s390/pci: implement minimal PCI error recovery")
---
 arch/s390/pci/pci_event.c |    4 ++++
 1 file changed, 4 insertions(+)
arch/s390/pci/pci_event.c +4 -0
@@ -106,6 +106,10 @@ static pci_ers_result_t zpci_event_do_er
 	struct zpci_dev *zdev = to_zpci(pdev);
 	int rc;
 
+	/* The underlying device may have been disabled by the event */
+	if (!zdev_enabled(zdev))
+		return PCI_ERS_RESULT_NEED_RESET;
+
 	pr_info("%s: Unblocking device access for examination\n", pci_name(pdev));
 	rc = zpci_reset_load_store_blocked(zdev);
 	if (rc) {
author maintainer
#34 lore raw

PATCH 34/178 firmware: arm_ffa: Move memory allocation outside the mutex locking

In reply to Greg Kroah-Hartman

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

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

From: Sudeep Holla <[email protected]>

[ Upstream commit 27e850c88df0e25474a8caeb2903e2e90b62c1dc ]

The notifier callback node allocation is currently done while holding the notify_lock mutex. While this is safe even if memory allocation may sleep, we need to move the allocation outside the locked region in preparation to move from using muxtes to rwlocks.

Move the memory allocation to avoid potential sleeping in atomic context once the locks are moved from mutex to rwlocks.

Fixes e0573444edbf ("firmware: arm_ffa: Add interfaces to request notification callbacks")

Message-Id: <[email protected]>

---
 drivers/firmware/arm_ffa/driver.c | 48 +++++++++++++++----------------
 1 file changed, 24 insertions(+), 24 deletions(-)
drivers/firmware/arm_ffa/driver.c +24 -24
index 6f75cdf297209..44eecb786e67b 100644
@@ -1250,13 +1250,12 @@ notifier_hnode_get_by_type(u16 notify_id, enum notify_type type)
 	return NULL;
 }
 
-static int
-update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb,
-		   void *cb_data, bool is_registration, bool is_framework)
+static int update_notifier_cb(struct ffa_device *dev, int notify_id,
+			      struct notifier_cb_info *cb, bool is_framework)
 {
 	struct notifier_cb_info *cb_info = NULL;
 	enum notify_type type = ffa_notify_type_get(dev->vm_id);
-	bool cb_found;
+	bool cb_found, is_registration = !!cb;
 
 	if (is_framework)
 		cb_info = notifier_hnode_get_by_vmid_uuid(notify_id, dev->vm_id,
@@ -1270,18 +1269,7 @@ update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb,
 		return -EINVAL;
 
 	if (is_registration) {
-		cb_info = kzalloc(sizeof(*cb_info), GFP_KERNEL);
-		if (!cb_info)
-			return -ENOMEM;
-
-		cb_info->dev = dev;
-		cb_info->cb_data = cb_data;
-		if (is_framework)
-			cb_info->fwk_cb = cb;
-		else
-			cb_info->cb = cb;
-
-		hash_add(drv_info->notifier_hash, &cb_info->hnode, notify_id);
+		hash_add(drv_info->notifier_hash, &cb->hnode, notify_id);
 	} else {
 		hash_del(&cb_info->hnode);
 		kfree(cb_info);
@@ -1303,8 +1291,7 @@ static int __ffa_notify_relinquish(struct ffa_device *dev, int notify_id,
 
 	mutex_lock(&drv_info->notify_lock);
 
-	rc = update_notifier_cb(dev, notify_id, NULL, NULL, false,
-				is_framework);
+	rc = update_notifier_cb(dev, notify_id, NULL, is_framework);
 	if (rc) {
 		pr_err("Could not unregister notification callback\n");
 		mutex_unlock(&drv_info->notify_lock);
@@ -1335,6 +1322,7 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu,
 {
 	int rc;
 	u32 flags = 0;
+	struct notifier_cb_info *cb_info = NULL;
 
 	if (ffa_notifications_disabled())
 		return -EOPNOTSUPP;
@@ -1342,6 +1330,17 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu,
 	if (notify_id >= FFA_MAX_NOTIFICATIONS)
 		return -EINVAL;
 
+	cb_info = kzalloc(sizeof(*cb_info), GFP_KERNEL);
+	if (!cb_info)
+		return -ENOMEM;
+
+	cb_info->dev = dev;
+	cb_info->cb_data = cb_data;
+	if (is_framework)
+		cb_info->fwk_cb = cb;
+	else
+		cb_info->cb = cb;
+
 	mutex_lock(&drv_info->notify_lock);
 
 	if (!is_framework) {
@@ -1349,21 +1348,22 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu,
 			flags = PER_VCPU_NOTIFICATION_FLAG;
 
 		rc = ffa_notification_bind(dev->vm_id, BIT(notify_id), flags);
-		if (rc) {
-			mutex_unlock(&drv_info->notify_lock);
-			return rc;
-		}
+		if (rc)
+			goto out_unlock_free;
 	}
 
-	rc = update_notifier_cb(dev, notify_id, cb, cb_data, true,
-				is_framework);
+	rc = update_notifier_cb(dev, notify_id, cb_info, is_framework);
 	if (rc) {
 		pr_err("Failed to register callback for %d - %d\n",
 		       notify_id, rc);
 		if (!is_framework)
 			ffa_notification_unbind(dev->vm_id, BIT(notify_id));
 	}
+
+out_unlock_free:
 	mutex_unlock(&drv_info->notify_lock);
+	if (rc)
+		kfree(cb_info);
 
 	return rc;
 }
Signature · 1 line
2.39.5
author maintainer
#35 lore raw

PATCH 35/178 firmware: arm_ffa: Replace mutex with rwlock to avoid sleep in atomic context

In reply to Greg Kroah-Hartman

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

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

From: Sudeep Holla <[email protected]>

[ Upstream commit 9ca7a421229bbdfbe2e1e628cff5cfa782720a10 ]

The current use of a mutex to protect the notifier hashtable accesses can lead to issues in the atomic context. It results in the below kernel warnings:

  |  BUG: sleeping function called from invalid context at kernel/locking/mutex.c:258
  |  in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 9, name: kworker/0:0
  |  preempt_count: 1, expected: 0
  |  RCU nest depth: 0, expected: 0
  |  CPU: 0 UID: 0 PID: 9 Comm: kworker/0:0 Not tainted 6.14.0 #4
  |  Workqueue: ffa_pcpu_irq_notification notif_pcpu_irq_work_fn
  |  Call trace:
  |   show_stack+0x18/0x24 (C)
  |   dump_stack_lvl+0x78/0x90
  |   dump_stack+0x18/0x24
  |   __might_resched+0x114/0x170
  |   __might_sleep+0x48/0x98
  |   mutex_lock+0x24/0x80
  |   handle_notif_callbacks+0x54/0xe0
  |   notif_get_and_handle+0x40/0x88
  |   generic_exec_single+0x80/0xc0
  |   smp_call_function_single+0xfc/0x1a0
  |   notif_pcpu_irq_work_fn+0x2c/0x38
  |   process_one_work+0x14c/0x2b4
  |   worker_thread+0x2e4/0x3e0
  |   kthread+0x13c/0x210
  |   ret_from_fork+0x10/0x20

To address this, replace the mutex with an rwlock to protect the notifier hashtable accesses. This ensures that read-side locking does not sleep and multiple readers can acquire the lock concurrently, avoiding unnecessary contention and potential deadlocks. Writer access remains exclusive, preserving correctness.

This change resolves warnings from lockdep about potential sleep in atomic context.

Cc Jens Wiklander <[email protected]>
Fixes e0573444edbf ("firmware: arm_ffa: Add interfaces to request notification callbacks")

Message-Id: <[email protected]>

---
 drivers/firmware/arm_ffa/driver.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)
drivers/firmware/arm_ffa/driver.c +11 -11
index 44eecb786e67b..37eb2e6c2f9f4 100644
@@ -110,7 +110,7 @@ struct ffa_drv_info {
 	struct work_struct sched_recv_irq_work;
 	struct xarray partition_info;
 	DECLARE_HASHTABLE(notifier_hash, ilog2(FFA_MAX_NOTIFICATIONS));
-	struct mutex notify_lock; /* lock to protect notifier hashtable  */
+	rwlock_t notify_lock; /* lock to protect notifier hashtable  */
 };
 
 static struct ffa_drv_info *drv_info;
@@ -1289,19 +1289,19 @@ static int __ffa_notify_relinquish(struct ffa_device *dev, int notify_id,
 	if (notify_id >= FFA_MAX_NOTIFICATIONS)
 		return -EINVAL;
 
-	mutex_lock(&drv_info->notify_lock);
+	write_lock(&drv_info->notify_lock);
 
 	rc = update_notifier_cb(dev, notify_id, NULL, is_framework);
 	if (rc) {
 		pr_err("Could not unregister notification callback\n");
-		mutex_unlock(&drv_info->notify_lock);
+		write_unlock(&drv_info->notify_lock);
 		return rc;
 	}
 
 	if (!is_framework)
 		rc = ffa_notification_unbind(dev->vm_id, BIT(notify_id));
 
-	mutex_unlock(&drv_info->notify_lock);
+	write_unlock(&drv_info->notify_lock);
 
 	return rc;
 }
@@ -1341,7 +1341,7 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu,
 	else
 		cb_info->cb = cb;
 
-	mutex_lock(&drv_info->notify_lock);
+	write_lock(&drv_info->notify_lock);
 
 	if (!is_framework) {
 		if (is_per_vcpu)
@@ -1361,7 +1361,7 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu,
 	}
 
 out_unlock_free:
-	mutex_unlock(&drv_info->notify_lock);
+	write_unlock(&drv_info->notify_lock);
 	if (rc)
 		kfree(cb_info);
 
@@ -1407,9 +1407,9 @@ static void handle_notif_callbacks(u64 bitmap, enum notify_type type)
 		if (!(bitmap & 1))
 			continue;
 
-		mutex_lock(&drv_info->notify_lock);
+		read_lock(&drv_info->notify_lock);
 		cb_info = notifier_hnode_get_by_type(notify_id, type);
-		mutex_unlock(&drv_info->notify_lock);
+		read_unlock(&drv_info->notify_lock);
 
 		if (cb_info && cb_info->cb)
 			cb_info->cb(notify_id, cb_info->cb_data);
@@ -1447,9 +1447,9 @@ static void handle_fwk_notif_callbacks(u32 bitmap)
 
 	ffa_rx_release();
 
-	mutex_lock(&drv_info->notify_lock);
+	read_lock(&drv_info->notify_lock);
 	cb_info = notifier_hnode_get_by_vmid_uuid(notify_id, target, &uuid);
-	mutex_unlock(&drv_info->notify_lock);
+	read_unlock(&drv_info->notify_lock);
 
 	if (cb_info && cb_info->fwk_cb)
 		cb_info->fwk_cb(notify_id, cb_info->cb_data, buf);
@@ -1974,7 +1974,7 @@ static void ffa_notifications_setup(void)
 		goto cleanup;
 
 	hash_init(drv_info->notifier_hash);
-	mutex_init(&drv_info->notify_lock);
+	rwlock_init(&drv_info->notify_lock);
 
 	drv_info->notif_enabled = true;
 	return;
Signature · 1 line
2.39.5
author maintainer
#36 lore raw

PATCH 36/178 firmware: arm_ffa: Fix the missing entry in struct ffa_indirect_msg_hdr

In reply to Greg Kroah-Hartman

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

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

From: Viresh Kumar <[email protected]>

[ Upstream commit 4c46a471be12216347ba707f8eadadbf5d68e698 ]

As per the spec, one 32 bit reserved entry is missing here, add it.

Fixes 910cc1acc9b4 ("firmware: arm_ffa: Add support for passing UUID in FFA_MSG_SEND2")

Message-Id: <28a624fbf416975de4fbe08cfbf7c2db89cb630e.1748948911.git.viresh.kumar@linaro.org>

---
 include/linux/arm_ffa.h | 1 +
 1 file changed, 1 insertion(+)
include/linux/arm_ffa.h +1 -0
index 5bded24dc24fe..e1634897e159c 100644
@@ -283,6 +283,7 @@ struct ffa_indirect_msg_hdr {
 	u32 offset;
 	u32 send_recv_id;
 	u32 size;
+	u32 res1;
 	uuid_t uuid;
 };
 
Signature · 1 line
2.39.5
author maintainer
#8 lore raw

PATCH 8/178 dt-bindings: net: sophgo,sg2044-dwmac: Drop status from the example

In reply to Greg Kroah-Hartman

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

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

From: Krzysztof Kozlowski <[email protected]>

commit f030713e5abf67d0a88864c8855f809c763af954 upstream.

Examples should be complete and should not have a 'status' property, especially a disabled one because this disables the dt_binding_check of the example against the schema. Dropping 'status' property shows missing other properties - phy-mode and phy-handle.

---
 Documentation/devicetree/bindings/net/sophgo,sg2044-dwmac.yaml | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/net/sophgo,sg2044-dwmac.yaml +2 -1
index 4dd2dc9c678b..8afbd9ebd73f 100644
@@ -80,6 +80,8 @@ examples:
       interrupt-parent = <&intc>;
       interrupts = <296 IRQ_TYPE_LEVEL_HIGH>;
       interrupt-names = "macirq";
+      phy-handle = <&phy0>;
+      phy-mode = "rgmii-id";
       resets = <&rst 30>;
       reset-names = "stmmaceth";
       snps,multicast-filter-bins = <0>;
@@ -91,7 +93,6 @@ examples:
       snps,mtl-rx-config = <&gmac0_mtl_rx_setup>;
       snps,mtl-tx-config = <&gmac0_mtl_tx_setup>;
       snps,axi-config = <&gmac0_stmmac_axi_setup>;
-      status = "disabled";
 
       gmac0_mtl_rx_setup: rx-queues-config {
         snps,rx-queues-to-use = <8>;
Signature · 1 line
2.50.0
author maintainer
#9 lore raw

PATCH 9/178 net: txgbe: request MISC IRQ in ndo_open

In reply to Greg Kroah-Hartman

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

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

From: Jiawen Wu <[email protected]>

commit cc9f7f65cd2f31150b10e6956f1f0882e1bbae49 upstream.

Move the creating of irq_domain for MISC IRQ from .probe to .ndo_open, and free it in .ndo_stop, to maintain consistency with the queue IRQs. This it for subsequent adjustments to the IRQ vectors.

---
 drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c  |    2 +-
 drivers/net/ethernet/wangxun/txgbe/txgbe_main.c |   22 ++++++++++------------
 2 files changed, 11 insertions(+), 13 deletions(-)
drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c +1 -1
@@ -68,7 +68,6 @@ free_queue_irqs:
 		free_irq(wx->msix_q_entries[vector].vector,
 			 wx->q_vector[vector]);
 	}
-	wx_reset_interrupt_capability(wx);
 	return err;
 }
 
@@ -172,6 +171,7 @@ void txgbe_free_misc_irq(struct txgbe *t
 	free_irq(txgbe->link_irq, txgbe);
 	free_irq(txgbe->misc.irq, txgbe);
 	txgbe_del_irq_domain(txgbe);
+	txgbe->wx->misc_irq_domain = false;
 }
 
 int txgbe_setup_misc_irq(struct txgbe *txgbe)
drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +10 -12
@@ -354,10 +354,14 @@ static int txgbe_open(struct net_device
 
 	wx_configure(wx);
 
-	err = txgbe_request_queue_irqs(wx);
+	err = txgbe_setup_misc_irq(wx->priv);
 	if (err)
 		goto err_free_resources;
 
+	err = txgbe_request_queue_irqs(wx);
+	if (err)
+		goto err_free_misc_irq;
+
 	/* Notify the stack of the actual queue counts. */
 	err = netif_set_real_num_tx_queues(netdev, wx->num_tx_queues);
 	if (err)
@@ -375,6 +379,9 @@ static int txgbe_open(struct net_device
 
 err_free_irq:
 	wx_free_irq(wx);
+err_free_misc_irq:
+	txgbe_free_misc_irq(wx->priv);
+	wx_reset_interrupt_capability(wx);
 err_free_resources:
 	wx_free_resources(wx);
 err_reset:
@@ -415,6 +422,7 @@ static int txgbe_close(struct net_device
 	wx_ptp_stop(wx);
 	txgbe_down(wx);
 	wx_free_irq(wx);
+	txgbe_free_misc_irq(wx->priv);
 	wx_free_resources(wx);
 	txgbe_fdir_filter_exit(wx);
 	wx_control_hw(wx, false);
@@ -460,7 +468,6 @@ static void txgbe_shutdown(struct pci_de
 int txgbe_setup_tc(struct net_device *dev, u8 tc)
 {
 	struct wx *wx = netdev_priv(dev);
-	struct txgbe *txgbe = wx->priv;
 
 	/* Hardware has to reinitialize queues and interrupts to
 	 * match packet buffer alignment. Unfortunately, the
@@ -471,7 +478,6 @@ int txgbe_setup_tc(struct net_device *de
 	else
 		txgbe_reset(wx);
 
-	txgbe_free_misc_irq(txgbe);
 	wx_clear_interrupt_scheme(wx);
 
 	if (tc)
@@ -480,7 +486,6 @@ int txgbe_setup_tc(struct net_device *de
 		netdev_reset_tc(dev);
 
 	wx_init_interrupt_scheme(wx);
-	txgbe_setup_misc_irq(txgbe);
 
 	if (netif_running(dev))
 		txgbe_open(dev);
@@ -729,13 +734,9 @@ static int txgbe_probe(struct pci_dev *p
 
 	txgbe_init_fdir(txgbe);
 
-	err = txgbe_setup_misc_irq(txgbe);
-	if (err)
-		goto err_release_hw;
-
 	err = txgbe_init_phy(txgbe);
 	if (err)
-		goto err_free_misc_irq;
+		goto err_release_hw;
 
 	err = register_netdev(netdev);
 	if (err)
@@ -763,8 +764,6 @@ static int txgbe_probe(struct pci_dev *p
 
 err_remove_phy:
 	txgbe_remove_phy(txgbe);
-err_free_misc_irq:
-	txgbe_free_misc_irq(txgbe);
 err_release_hw:
 	wx_clear_interrupt_scheme(wx);
 	wx_control_hw(wx, false);
@@ -798,7 +797,6 @@ static void txgbe_remove(struct pci_dev
 	unregister_netdev(netdev);
 
 	txgbe_remove_phy(txgbe);
-	txgbe_free_misc_irq(txgbe);
 	wx_free_isb_resources(wx);
 
 	pci_release_selected_regions(pdev,
author maintainer
#10 lore raw

PATCH 10/178 vsock/vmci: Clear the vmci transport packet properly when initializing it

In reply to Greg Kroah-Hartman

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

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

From: HarshaVardhana S A <[email protected]>

commit 223e2288f4b8c262a864e2c03964ffac91744cd5 upstream.

In vmci_transport_packet_init memset the vmci_transport_packet before populating the fields to avoid any uninitialised data being left in the structure.

Cc Bryan Tan <[email protected]>
Cc Vishnu Dasa <[email protected]>
Cc Broadcom internal kernel review list
Cc Stefano Garzarella <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Eric Dumazet <[email protected]>
Cc Jakub Kicinski <[email protected]>
Cc Paolo Abeni <[email protected]>
Cc Simon Horman <[email protected]>
Cc stable <[email protected]>
Fixes d021c344051a ("VSOCK: Introduce VM Sockets")
---
 net/vmw_vsock/vmci_transport.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/vmw_vsock/vmci_transport.c +2 -2
@@ -119,6 +119,8 @@ vmci_transport_packet_init(struct vmci_t
 			   u16 proto,
 			   struct vmci_handle handle)
 {
+	memset(pkt, 0, sizeof(*pkt));
+
 	/* We register the stream control handler as an any cid handle so we
 	 * must always send from a source address of VMADDR_CID_ANY
 	 */
@@ -131,8 +133,6 @@ vmci_transport_packet_init(struct vmci_t
 	pkt->type = type;
 	pkt->src_port = src->svm_port;
 	pkt->dst_port = dst->svm_port;
-	memset(&pkt->proto, 0, sizeof(pkt->proto));
-	memset(&pkt->_reserved2, 0, sizeof(pkt->_reserved2));
 
 	switch (pkt->type) {
 	case VMCI_TRANSPORT_PACKET_TYPE_INVALID:
author maintainer
#11 lore raw

PATCH 11/178 iommufd/selftest: Add missing close(mfd) in memfd_mmap()

In reply to Greg Kroah-Hartman

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

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

From: Nicolin Chen <[email protected]>

commit 4b75e3babb85238912f50bbe0647bae08242a9b6 upstream.

Do not forget to close mfd in the error paths, since none of the callers would close it when ASSERT_NE(MAP_FAILED, buf) fails.

---
 tools/testing/selftests/iommu/iommufd_utils.h |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
tools/testing/selftests/iommu/iommufd_utils.h +7 -2
@@ -60,13 +60,18 @@ static inline void *memfd_mmap(size_t le
 {
 	int mfd_flags = (flags & MAP_HUGETLB) ? MFD_HUGETLB : 0;
 	int mfd = memfd_create("buffer", mfd_flags);
+	void *buf = MAP_FAILED;
 
 	if (mfd <= 0)
 		return MAP_FAILED;
 	if (ftruncate(mfd, length))
-		return MAP_FAILED;
+		goto out;
 	*mfd_p = mfd;
-	return mmap(0, length, prot, flags, mfd, 0);
+	buf = mmap(0, length, prot, flags, mfd, 0);
+out:
+	if (buf == MAP_FAILED)
+		close(mfd);
+	return buf;
 }
 
 /*
author maintainer
#12 lore raw

PATCH 12/178 iommufd/selftest: Add asserts testing global mfd

In reply to Greg Kroah-Hartman

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

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

From: Nicolin Chen <[email protected]>

commit a9bf67ee170514b17541038c60bb94cb2cf5732f upstream.

The mfd and mfd_buffer will be used in the tests directly without an extra check. Test them in setup_sizes() to ensure they are safe to use.

---
 tools/testing/selftests/iommu/iommufd.c |    2 ++
 1 file changed, 2 insertions(+)
tools/testing/selftests/iommu/iommufd.c +2 -0
@@ -54,6 +54,8 @@ static __attribute__((constructor)) void
 
 	mfd_buffer = memfd_mmap(BUFFER_SIZE, PROT_READ | PROT_WRITE, MAP_SHARED,
 				&mfd);
+	assert(mfd_buffer != MAP_FAILED);
+	assert(mfd > 0);
 }
 
 FIXTURE(iommufd)
author maintainer
#13 lore raw

PATCH 13/178 net: libwx: fix the incorrect display of the queue number

In reply to Greg Kroah-Hartman

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

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

From: Jiawen Wu <[email protected]>

commit 5186ff7e1d0e26aaef998ba18b31c79c28d1441f upstream.

When setting "ethtool -L eth0 combined 1", the number of RX/TX queue is changed to be 1. RSS is disabled at this moment, and the indices of FDIR have not be changed in wx_set_rss_queues(). So the combined count still shows the previous value. This issue was introduced when supporting FDIR. Fix it for those devices that support FDIR.

---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -0
@@ -1641,6 +1641,7 @@ static void wx_set_rss_queues(struct wx
 
 	clear_bit(WX_FLAG_FDIR_HASH, wx->flags);
 
+	wx->ring_feature[RING_F_FDIR].indices = 1;
 	/* Use Flow Director in addition to RSS to ensure the best
 	 * distribution of flows across cores, even when an FDIR flow
 	 * isn't matched.
author maintainer
#4 lore raw

PATCH 4/178 virtio-net: xsk: rx: fix the frames length check

In reply to Greg Kroah-Hartman

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

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

From: Bui Quang Minh <[email protected]>

commit 5177373c31318c3c6a190383bfd232e6cf565c36 upstream.

When calling buf_to_xdp, the len argument is the frame data's length without virtio header's length (vi->hdr_len). We check that len with

	xsk_pool_get_rx_frame_size() + vi->hdr_len

to ensure the provided len does not larger than the allocated chunk size. The additional vi->hdr_len is because in virtnet_add_recvbuf_xsk, we use part of XDP_PACKET_HEADROOM for virtio header and ask the vhost to start placing data from

	hard_start + XDP_PACKET_HEADROOM - vi->hdr_len

not
hard_start + XDP_PACKET_HEADROOM

But the first buffer has virtio_header, so the maximum frame's length in the first buffer can only be

xsk_pool_get_rx_frame_size()
not

	xsk_pool_get_rx_frame_size() + vi->hdr_len

like in the current check.

This commit adds an additional argument to buf_to_xdp differentiate between the first buffer and other ones to correctly calculate the maximum frame's length.

---
 drivers/net/virtio_net.c |   22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)
drivers/net/virtio_net.c +18 -4
@@ -1127,15 +1127,29 @@ static void check_sq_full_and_disable(st
 	}
 }
 
+/* Note that @len is the length of received data without virtio header */
 static struct xdp_buff *buf_to_xdp(struct virtnet_info *vi,
-				   struct receive_queue *rq, void *buf, u32 len)
+				   struct receive_queue *rq, void *buf,
+				   u32 len, bool first_buf)
 {
 	struct xdp_buff *xdp;
 	u32 bufsize;
 
 	xdp = (struct xdp_buff *)buf;
 
-	bufsize = xsk_pool_get_rx_frame_size(rq->xsk_pool) + vi->hdr_len;
+	/* In virtnet_add_recvbuf_xsk, we use part of XDP_PACKET_HEADROOM for
+	 * virtio header and ask the vhost to fill data from
+	 *         hard_start + XDP_PACKET_HEADROOM - vi->hdr_len
+	 * The first buffer has virtio header so the remaining region for frame
+	 * data is
+	 *         xsk_pool_get_rx_frame_size()
+	 * While other buffers than the first one do not have virtio header, so
+	 * the maximum frame data's length can be
+	 *         xsk_pool_get_rx_frame_size() + vi->hdr_len
+	 */
+	bufsize = xsk_pool_get_rx_frame_size(rq->xsk_pool);
+	if (!first_buf)
+		bufsize += vi->hdr_len;
 
 	if (unlikely(len > bufsize)) {
 		pr_debug("%s: rx error: len %u exceeds truesize %u\n",
@@ -1260,7 +1274,7 @@ static int xsk_append_merge_buffer(struc
 
 		u64_stats_add(&stats->bytes, len);
 
-		xdp = buf_to_xdp(vi, rq, buf, len);
+		xdp = buf_to_xdp(vi, rq, buf, len, false);
 		if (!xdp)
 			goto err;
 
@@ -1358,7 +1372,7 @@ static void virtnet_receive_xsk_buf(stru
 
 	u64_stats_add(&stats->bytes, len);
 
-	xdp = buf_to_xdp(vi, rq, buf, len);
+	xdp = buf_to_xdp(vi, rq, buf, len, true);
 	if (!xdp)
 		return;
 
author maintainer
#38 lore raw

PATCH 38/178 RDMA/mlx5: reduce stack usage in mlx5_ib_ufile_hw_cleanup

In reply to Greg Kroah-Hartman

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

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

From: Arnd Bergmann <[email protected]>

[ Upstream commit b26852daaa83f535109253d114426d1fa674155d ]

This function has an array of eight mlx5_async_cmd structures, which often fits on the stack, but depending on the configuration can end up blowing the stack frame warning limit:

drivers/infiniband/hw/mlx5/devx.c:2670:6: error: stack frame size (1392) exceeds limit (1280) in 'mlx5_ib_ufile_hw_cleanup' [-Werror,-Wframe-larger-than]

Change this to a dynamic allocation instead. While a kmalloc() can theoretically fail, a GFP_KERNEL allocation under a page will block until memory has been freed up, so in the worst case, this only adds extra time in an already constrained environment.

Fixes 7c891a4dbcc1 ("RDMA/mlx5: Add implementation for ufile_hw_cleanup device operation")
---
 drivers/infiniband/hw/mlx5/devx.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/infiniband/hw/mlx5/devx.c +7 -1
index 2479da8620ca9..bceae1c1f9801 100644
@@ -2669,7 +2669,7 @@ static void devx_wait_async_destroy(struct mlx5_async_cmd *cmd)
 
 void mlx5_ib_ufile_hw_cleanup(struct ib_uverbs_file *ufile)
 {
-	struct mlx5_async_cmd async_cmd[MAX_ASYNC_CMDS];
+	struct mlx5_async_cmd *async_cmd;
 	struct ib_ucontext *ucontext = ufile->ucontext;
 	struct ib_device *device = ucontext->device;
 	struct mlx5_ib_dev *dev = to_mdev(device);
@@ -2678,6 +2678,10 @@ void mlx5_ib_ufile_hw_cleanup(struct ib_uverbs_file *ufile)
 	int head = 0;
 	int tail = 0;
 
+	async_cmd = kcalloc(MAX_ASYNC_CMDS, sizeof(*async_cmd), GFP_KERNEL);
+	if (!async_cmd)
+		return;
+
 	list_for_each_entry(uobject, &ufile->uobjects, list) {
 		WARN_ON(uverbs_try_lock_object(uobject, UVERBS_LOOKUP_WRITE));
 
@@ -2713,6 +2717,8 @@ void mlx5_ib_ufile_hw_cleanup(struct ib_uverbs_file *ufile)
 		devx_wait_async_destroy(&async_cmd[head % MAX_ASYNC_CMDS]);
 		head++;
 	}
+
+	kfree(async_cmd);
 }
 
 static ssize_t devx_async_cmd_event_read(struct file *filp, char __user *buf,
Signature · 1 line
2.39.5
author maintainer
#47 lore raw

PATCH 47/178 nfs: Clean up /proc/net/rpc/nfs when nfs_fs_proc_net_init() fails.

In reply to Greg Kroah-Hartman

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

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

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit e8d6f3ab59468e230f3253efe5cb63efa35289f7 ]

syzbot reported a warning below [1] following a fault injection in nfs_fs_proc_net_init(). [0]

When nfs_fs_proc_net_init() fails, /proc/net/rpc/nfs is not removed.

Later, rpc_proc_exit() tries to remove /proc/net/rpc, and the warning is logged as the directory is not empty.

Let's handle the error of nfs_fs_proc_net_init() properly.

[0]:
FAULT_INJECTION: forcing a failure.
name failslab, interval 1, probability 0, space 0, times 0

CPU: 1 UID: 0 PID: 6120 Comm: syz.2.27 Not tainted 6.16.0-rc1-syzkaller-00010-g2c4a1f3fe03e #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
Call Trace:
 <TASK>
  dump_stack_lvl (lib/dump_stack.c:123)
 should_fail_ex (lib/fault-inject.c:73 lib/fault-inject.c:174)
 should_failslab (mm/failslab.c:46)
 kmem_cache_alloc_noprof (mm/slub.c:4178 mm/slub.c:4204)
 __proc_create (fs/proc/generic.c:427)
 proc_create_reg (fs/proc/generic.c:554)
 proc_create_net_data (fs/proc/proc_net.c:120)
 nfs_fs_proc_net_init (fs/nfs/client.c:1409)
 nfs_net_init (fs/nfs/inode.c:2600)
 ops_init (net/core/net_namespace.c:138)
 setup_net (net/core/net_namespace.c:443)
 copy_net_ns (net/core/net_namespace.c:576)
 create_new_namespaces (kernel/nsproxy.c:110)
 unshare_nsproxy_namespaces (kernel/nsproxy.c:218 (discriminator 4))
 ksys_unshare (kernel/fork.c:3123)
 __x64_sys_unshare (kernel/fork.c:3190)
 do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94)
 entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
 </TASK>

[1]:
remove_proc_entry: removing non-empty directory 'net/rpc', leaking at least 'nfs'

 WARNING: CPU: 1 PID: 6120 at fs/proc/generic.c:727 remove_proc_entry+0x45e/0x530 fs/proc/generic.c:727
Modules linked in:
CPU: 1 UID: 0 PID: 6120 Comm: syz.2.27 Not tainted 6.16.0-rc1-syzkaller-00010-g2c4a1f3fe03e #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
 RIP: 0010:remove_proc_entry+0x45e/0x530 fs/proc/generic.c:727
Code: 3c 02 00 0f 85 85 00 00 00 48 8b 93 d8 00 00 00 4d 89 f0 4c 89 e9 48 c7 c6 40 ba a2 8b 48 c7 c7 60 b9 a2 8b e8 33 81 1d ff 90 <0f> 0b 90 90 e9 5f fe ff ff e8 04 69 5e ff 90 48 b8 00 00 00 00 00
RSP: 0018:ffffc90003637b08 EFLAGS: 00010282
RAX: 0000000000000000 RBX: ffff88805f534140 RCX: ffffffff817a92c8
RDX: ffff88807da99e00 RSI: ffffffff817a92d5 RDI: 0000000000000001
RBP: ffff888033431ac0 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000001 R12: ffff888033431a00
R13: ffff888033431ae4 R14: ffff888033184724 R15: dffffc0000000000
FS:  0000555580328500(0000) GS:ffff888124a62000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f71733743e0 CR3: 000000007f618000 CR4: 00000000003526f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
  sunrpc_exit_net+0x46/0x90 net/sunrpc/sunrpc_syms.c:76
  ops_exit_list net/core/net_namespace.c:200 [inline]
  ops_undo_list+0x2eb/0xab0 net/core/net_namespace.c:253
  setup_net+0x2e1/0x510 net/core/net_namespace.c:457
  copy_net_ns+0x2a6/0x5f0 net/core/net_namespace.c:574
  create_new_namespaces+0x3ea/0xa90 kernel/nsproxy.c:110
  unshare_nsproxy_namespaces+0xc0/0x1f0 kernel/nsproxy.c:218
  ksys_unshare+0x45b/0xa40 kernel/fork.c:3121
  __do_sys_unshare kernel/fork.c:3192 [inline]
  __se_sys_unshare kernel/fork.c:3190 [inline]
  __x64_sys_unshare+0x31/0x40 kernel/fork.c:3190
  do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
  do_syscall_64+0xcd/0x490 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7fa1a6b8e929
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fff3a090368 EFLAGS: 00000246 ORIG_RAX: 0000000000000110
RAX: ffffffffffffffda RBX: 00007fa1a6db5fa0 RCX: 00007fa1a6b8e929
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000040000080
RBP: 00007fa1a6c10b39 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007fa1a6db5fa0 R14: 00007fa1a6db5fa0 R15: 0000000000000001
 </TASK>
Fixes d47151b79e32 ("nfs: expose /proc/net/sunrpc/nfs in net namespaces")
Reported-by [email protected]
---
 fs/nfs/inode.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)
fs/nfs/inode.c +14 -3
index 8ab7868807a7d..a2fa6bc4d74e3 100644
@@ -2589,15 +2589,26 @@ EXPORT_SYMBOL_GPL(nfs_net_id);
 static int nfs_net_init(struct net *net)
 {
 	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	int err;
 
 	nfs_clients_init(net);
 
 	if (!rpc_proc_register(net, &nn->rpcstats)) {
-		nfs_clients_exit(net);
-		return -ENOMEM;
+		err = -ENOMEM;
+		goto err_proc_rpc;
 	}
 
-	return nfs_fs_proc_net_init(net);
+	err = nfs_fs_proc_net_init(net);
+	if (err)
+		goto err_proc_nfs;
+
+	return 0;
+
+err_proc_nfs:
+	rpc_proc_unregister(net, "nfs");
+err_proc_rpc:
+	nfs_clients_exit(net);
+	return err;
 }
 
 static void nfs_net_exit(struct net *net)
Signature · 1 line
2.39.5
author maintainer
#48 lore raw

PATCH 48/178 NFSv4/pNFS: Fix a race to wake on NFS_LAYOUT_DRAIN

In reply to Greg Kroah-Hartman

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

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

From: Benjamin Coddington <[email protected]>

[ Upstream commit c01776287414ca43412d1319d2877cbad65444ac ]

We found a few different systems hung up in writeback waiting on the same page lock, and one task waiting on the NFS_LAYOUT_DRAIN bit in pnfs_update_layout(), however the pnfs_layout_hdr's plh_outstanding count was zero.

It seems most likely that this is another race between the waiter and waker similar to commit ed0172af5d6f ("SUNRPC: Fix a race to wake a sync task"). Fix it up by applying the advised barrier.

Fixes 880265c77ac4 ("pNFS: Avoid a live lock condition in pnfs_update_layout()")
---
 fs/nfs/pnfs.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
fs/nfs/pnfs.c +3 -1
index 3adb7d0dbec7a..1a7ec68bde153 100644
@@ -2059,8 +2059,10 @@ static void nfs_layoutget_begin(struct pnfs_layout_hdr *lo)
 static void nfs_layoutget_end(struct pnfs_layout_hdr *lo)
 {
 	if (atomic_dec_and_test(&lo->plh_outstanding) &&
-	    test_and_clear_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags))
+	    test_and_clear_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags)) {
+		smp_mb__after_atomic();
 		wake_up_bit(&lo->plh_flags, NFS_LAYOUT_DRAIN);
+	}
 }
 
 static bool pnfs_is_first_layoutget(struct pnfs_layout_hdr *lo)
Signature · 1 line
2.39.5
author maintainer
#49 lore raw

PATCH 49/178 scsi: qla2xxx: Fix DMA mapping test in qla24xx_get_port_database()

In reply to Greg Kroah-Hartman

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

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

From: Thomas Fourier <[email protected]>

[ Upstream commit c3b214719a87735d4f67333a8ef3c0e31a34837c ]

dma_map_XXX() functions return as error values DMA_MAPPING_ERROR which is often ~0. The error value should be tested with dma_mapping_error() like it was done in qla26xx_dport_diagnostics().

Fixes 818c7f87a177 ("scsi: qla2xxx: Add changes in preparation for vendor extended FDMI/RDP")
---
 drivers/scsi/qla2xxx/qla_mbx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_mbx.c +1 -1
index 0cd6f3e148824..13b6cb1b93acd 100644
@@ -2147,7 +2147,7 @@ qla24xx_get_port_database(scsi_qla_host_t *vha, u16 nport_handle,
 
 	pdb_dma = dma_map_single(&vha->hw->pdev->dev, pdb,
 	    sizeof(*pdb), DMA_FROM_DEVICE);
-	if (!pdb_dma) {
+	if (dma_mapping_error(&vha->hw->pdev->dev, pdb_dma)) {
 		ql_log(ql_log_warn, vha, 0x1116, "Failed to map dma buffer.\n");
 		return QLA_MEMORY_ALLOC_FAILED;
 	}
Signature · 1 line
2.39.5
author maintainer
#50 lore raw

PATCH 50/178 scsi: qla4xxx: Fix missing DMA mapping error in qla4xxx_alloc_pdu()

In reply to Greg Kroah-Hartman

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

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

From: Thomas Fourier <[email protected]>

[ Upstream commit 00f452a1b084efbe8dcb60a29860527944a002a1 ]

dma_map_XXX() can fail and should be tested for errors with dma_mapping_error().

Fixes b3a271a94d00 ("[SCSI] qla4xxx: support iscsiadm session mgmt")
---
 drivers/scsi/qla4xxx/ql4_os.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/scsi/qla4xxx/ql4_os.c +2 -0
index d540d66e6ffc4..80c4b4e7526b1 100644
@@ -3420,6 +3420,8 @@ static int qla4xxx_alloc_pdu(struct iscsi_task *task, uint8_t opcode)
 		task_data->data_dma = dma_map_single(&ha->pdev->dev, task->data,
 						     task->data_count,
 						     DMA_TO_DEVICE);
+		if (dma_mapping_error(&ha->pdev->dev, task_data->data_dma))
+			return -ENOMEM;
 	}
 
 	DEBUG2(ql4_printk(KERN_INFO, ha, "%s: MaxRecvLen %u, iscsi hrd %d\n",
Signature · 1 line
2.39.5
author maintainer
#51 lore raw

PATCH 51/178 scsi: sd: Fix VPD page 0xb7 length check

In reply to Greg Kroah-Hartman

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

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

From: jackysliu <[email protected]>

[ Upstream commit 8889676cd62161896f1d861ce294adc29c4f2cb5 ]

sd_read_block_limits_ext() currently assumes that vpd->len excludes the size of the page header. However, vpd->len describes the size of the entire VPD page, therefore the sanity check is incorrect.

In practice this is not really a problem since we don't attach VPD pages unless they actually report data trailing the header. But fix the length check regardless.

This issue was identified by Wukong-Agent (formerly Tencent Woodpecker), a code security AI agent, through static code analysis.

[mkp: rewrote patch description]

Fixes 96b171d6dba6 ("scsi: core: Query the Block Limits Extension VPD page")
---
 drivers/scsi/sd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/sd.c +1 -1
index 950d8c9fb8843..89d5c4b17bc46 100644
@@ -3384,7 +3384,7 @@ static void sd_read_block_limits_ext(struct scsi_disk *sdkp)
 
 	rcu_read_lock();
 	vpd = rcu_dereference(sdkp->device->vpd_pgb7);
-	if (vpd && vpd->len >= 2)
+	if (vpd && vpd->len >= 6)
 		sdkp->rscs = vpd->data[5] & 1;
 	rcu_read_unlock();
 }
Signature · 1 line
2.39.5
author maintainer
#52 lore raw

PATCH 52/178 scsi: core: Enforce unlimited max_segment_size when virt_boundary_mask is set

In reply to Greg Kroah-Hartman

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

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

From: Christoph Hellwig <[email protected]>

[ Upstream commit 4937e604ca24c41cae3296d069c871c2f3f519c8 ]

The virt_boundary_mask limit requires an unlimited max_segment_size for bio splitting to not corrupt data. Historically, the block layer tried to validate this, although the check was half-hearted until the addition of the atomic queue limits API. The full blown check then triggered issues with stacked devices incorrectly inheriting limits such as the virt boundary and got disabled in commit b561ea56a264 ("block: allow device to have both virt_boundary_mask and max segment size") instead of fixing the issue properly.

Ensure that the SCSI mid layer doesn't set the default low max_segment_size limit for this case, and check for invalid max_segment_size values in the host template, similar to the original block layer check given that SCSI devices can't be stacked.

This fixes reported data corruption on storvsc, although as far as I can tell storvsc always failed to properly set the max_segment_size limit as the SCSI APIs historically applied that when setting up the host, while storvsc only set the virt_boundary_mask when configuring the scsi_device.

Fixes 81988a0e6b03 ("storvsc: get rid of bounce buffer")
Fixes b561ea56a264 ("block: allow device to have both virt_boundary_mask and max segment size")
---
 drivers/scsi/hosts.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
drivers/scsi/hosts.c +11 -7
index e021f1106beab..cc5d05dc395c4 100644
@@ -473,10 +473,17 @@ struct Scsi_Host *scsi_host_alloc(const struct scsi_host_template *sht, int priv
 	else
 		shost->max_sectors = SCSI_DEFAULT_MAX_SECTORS;
 
-	if (sht->max_segment_size)
-		shost->max_segment_size = sht->max_segment_size;
-	else
-		shost->max_segment_size = BLK_MAX_SEGMENT_SIZE;
+	shost->virt_boundary_mask = sht->virt_boundary_mask;
+	if (shost->virt_boundary_mask) {
+		WARN_ON_ONCE(sht->max_segment_size &&
+			     sht->max_segment_size != UINT_MAX);
+		shost->max_segment_size = UINT_MAX;
+	} else {
+		if (sht->max_segment_size)
+			shost->max_segment_size = sht->max_segment_size;
+		else
+			shost->max_segment_size = BLK_MAX_SEGMENT_SIZE;
+	}
 
 	/* 32-byte (dword) is a common minimum for HBAs. */
 	if (sht->dma_alignment)
@@ -492,9 +499,6 @@ struct Scsi_Host *scsi_host_alloc(const struct scsi_host_template *sht, int priv
 	else
 		shost->dma_boundary = 0xffffffff;
 
-	if (sht->virt_boundary_mask)
-		shost->virt_boundary_mask = sht->virt_boundary_mask;
-
 	device_initialize(&shost->shost_gendev);
 	dev_set_name(&shost->shost_gendev, "host%d", shost->host_no);
 	shost->shost_gendev.bus = &scsi_bus_type;
Signature · 1 line
2.39.5
author maintainer
#53 lore raw

PATCH 53/178 scsi: ufs: core: Fix spelling of a sysfs attribute name

In reply to Greg Kroah-Hartman

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

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

From: Bart Van Assche <[email protected]>

[ Upstream commit 021f243627ead17eb6500170256d3d9be787dad8 ]

Change "resourse" into "resource" in the name of a sysfs attribute.

---
 Documentation/ABI/testing/sysfs-driver-ufs | 2 +-
 drivers/ufs/core/ufs-sysfs.c               | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
Documentation/ABI/testing/sysfs-driver-ufs +1 -1
index e36d2de16cbda..0397664e869a0 100644
@@ -711,7 +711,7 @@ Description:	This file shows the thin provisioning type. This is one of
 
 		The file is read only.
 
-What:		/sys/class/scsi_device/*/device/unit_descriptor/physical_memory_resourse_count
+What:		/sys/class/scsi_device/*/device/unit_descriptor/physical_memory_resource_count
 Date:		February 2018
 Contact:	Stanislav Nijnikov <[email protected]>
 Description:	This file shows the total physical memory resources. This is
drivers/ufs/core/ufs-sysfs.c +2 -2
index 634cf163f4cb1..c53f924f7fa83 100644
@@ -1675,7 +1675,7 @@ UFS_UNIT_DESC_PARAM(logical_block_size, _LOGICAL_BLK_SIZE, 1);
 UFS_UNIT_DESC_PARAM(logical_block_count, _LOGICAL_BLK_COUNT, 8);
 UFS_UNIT_DESC_PARAM(erase_block_size, _ERASE_BLK_SIZE, 4);
 UFS_UNIT_DESC_PARAM(provisioning_type, _PROVISIONING_TYPE, 1);
-UFS_UNIT_DESC_PARAM(physical_memory_resourse_count, _PHY_MEM_RSRC_CNT, 8);
+UFS_UNIT_DESC_PARAM(physical_memory_resource_count, _PHY_MEM_RSRC_CNT, 8);
 UFS_UNIT_DESC_PARAM(context_capabilities, _CTX_CAPABILITIES, 2);
 UFS_UNIT_DESC_PARAM(large_unit_granularity, _LARGE_UNIT_SIZE_M1, 1);
 UFS_UNIT_DESC_PARAM(wb_buf_alloc_units, _WB_BUF_ALLOC_UNITS, 4);
@@ -1692,7 +1692,7 @@ static struct attribute *ufs_sysfs_unit_descriptor[] = {
 	&dev_attr_logical_block_count.attr,
 	&dev_attr_erase_block_size.attr,
 	&dev_attr_provisioning_type.attr,
-	&dev_attr_physical_memory_resourse_count.attr,
+	&dev_attr_physical_memory_resource_count.attr,
 	&dev_attr_context_capabilities.attr,
 	&dev_attr_large_unit_granularity.attr,
 	&dev_attr_wb_buf_alloc_units.attr,
Signature · 1 line
2.39.5
author maintainer
#54 lore raw

PATCH 54/178 IB/mlx5: Fix potential deadlock in MR deregistration

In reply to Greg Kroah-Hartman

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

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

From: Or Har-Toov <[email protected]>

[ Upstream commit 2ed25aa7f7711f508b6120e336f05cd9d49943c0 ]

The issue arises when kzalloc() is invoked while holding umem_mutex or any other lock acquired under umem_mutex. This is problematic because kzalloc() can trigger fs_reclaim_aqcuire(), which may, in turn, invoke mmu_notifier_invalidate_range_start(). This function can lead to mlx5_ib_invalidate_range(), which attempts to acquire umem_mutex again, resulting in a deadlock.

The problematic flow:

             CPU0                      |              CPU1
---------------------------------------|------------------------------------------------
mlx5_ib_dereg_mr()                     |
 → revoke_mr()                         |
   → mutex_lock(&umem_odp->umem_mutex) |
                                       | mlx5_mkey_cache_init()
                                       |  → mutex_lock(&dev->cache.rb_lock)
                                       |  → mlx5r_cache_create_ent_locked()
                                       |    → kzalloc(GFP_KERNEL)
                                       |      → fs_reclaim()
                                       |        → mmu_notifier_invalidate_range_start()
                                       |          → mlx5_ib_invalidate_range()
                                       |            → mutex_lock(&umem_odp->umem_mutex)
   → cache_ent_find_and_store()        |
     → mutex_lock(&dev->cache.rb_lock) |

Additionally, when kzalloc() is called from within cache_ent_find_and_store(), we encounter the same deadlock due to re-acquisition of umem_mutex.

Solve by releasing umem_mutex in dereg_mr() after umr_revoke_mr() and before acquiring rb_lock. This ensures that we don't hold umem_mutex while performing memory allocations that could trigger the reclaim path.

This change prevents the deadlock by ensuring proper lock ordering and avoiding holding locks during memory allocation operations that could trigger the reclaim path.

The following lockdep warning demonstrates the deadlock:

python3/20557 is trying to acquire lock:
ffff888387542128 (&umem_odp->umem_mutex){+.+.}-{4:4}, at:
mlx5_ib_invalidate_range+0x5b/0x550 [mlx5_ib]

but task is already holding lock:
ffffffff82f6b840 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}, at:
unmap_vmas+0x7b/0x1a0

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}:

       fs_reclaim_acquire+0x60/0xd0
       mem_cgroup_css_alloc+0x6f/0x9b0
       cgroup_init_subsys+0xa4/0x240
       cgroup_init+0x1c8/0x510
       start_kernel+0x747/0x760
       x86_64_start_reservations+0x25/0x30
       x86_64_start_kernel+0x73/0x80
       common_startup_64+0x129/0x138

-> #2 (fs_reclaim){+.+.}-{0:0}:

       fs_reclaim_acquire+0x91/0xd0
       __kmalloc_cache_noprof+0x4d/0x4c0
       mlx5r_cache_create_ent_locked+0x75/0x620 [mlx5_ib]
       mlx5_mkey_cache_init+0x186/0x360 [mlx5_ib]
       mlx5_ib_stage_post_ib_reg_umr_init+0x3c/0x60 [mlx5_ib]
       __mlx5_ib_add+0x4b/0x190 [mlx5_ib]
       mlx5r_probe+0xd9/0x320 [mlx5_ib]
       auxiliary_bus_probe+0x42/0x70
       really_probe+0xdb/0x360
       __driver_probe_device+0x8f/0x130
       driver_probe_device+0x1f/0xb0
       __driver_attach+0xd4/0x1f0
       bus_for_each_dev+0x79/0xd0
       bus_add_driver+0xf0/0x200
       driver_register+0x6e/0xc0
       __auxiliary_driver_register+0x6a/0xc0
       do_one_initcall+0x5e/0x390
       do_init_module+0x88/0x240
       init_module_from_file+0x85/0xc0
       idempotent_init_module+0x104/0x300
       __x64_sys_finit_module+0x68/0xc0
       do_syscall_64+0x6d/0x140
       entry_SYSCALL_64_after_hwframe+0x4b/0x53

-> #1 (&dev->cache.rb_lock){+.+.}-{4:4}:

       __mutex_lock+0x98/0xf10
       __mlx5_ib_dereg_mr+0x6f2/0x890 [mlx5_ib]
       mlx5_ib_dereg_mr+0x21/0x110 [mlx5_ib]
       ib_dereg_mr_user+0x85/0x1f0 [ib_core]
       uverbs_free_mr+0x19/0x30 [ib_uverbs]
       destroy_hw_idr_uobject+0x21/0x80 [ib_uverbs]
       uverbs_destroy_uobject+0x60/0x3d0 [ib_uverbs]
       uobj_destroy+0x57/0xa0 [ib_uverbs]
       ib_uverbs_cmd_verbs+0x4d5/0x1210 [ib_uverbs]
       ib_uverbs_ioctl+0x129/0x230 [ib_uverbs]
       __x64_sys_ioctl+0x596/0xaa0
       do_syscall_64+0x6d/0x140
       entry_SYSCALL_64_after_hwframe+0x4b/0x53

-> #0 (&umem_odp->umem_mutex){+.+.}-{4:4}:

       __lock_acquire+0x1826/0x2f00
       lock_acquire+0xd3/0x2e0
       __mutex_lock+0x98/0xf10
       mlx5_ib_invalidate_range+0x5b/0x550 [mlx5_ib]
       __mmu_notifier_invalidate_range_start+0x18e/0x1f0
       unmap_vmas+0x182/0x1a0
       exit_mmap+0xf3/0x4a0
       mmput+0x3a/0x100
       do_exit+0x2b9/0xa90
       do_group_exit+0x32/0xa0
       get_signal+0xc32/0xcb0
       arch_do_signal_or_restart+0x29/0x1d0
       syscall_exit_to_user_mode+0x105/0x1d0
       do_syscall_64+0x79/0x140
       entry_SYSCALL_64_after_hwframe+0x4b/0x53

 Chain exists of:
 &dev->cache.rb_lock --> mmu_notifier_invalidate_range_start -->
 &umem_odp->umem_mutex

 Possible unsafe locking scenario:

CPU0 CPU1
---- ----

   lock(&umem_odp->umem_mutex);
                                lock(mmu_notifier_invalidate_range_start);
                                lock(&umem_odp->umem_mutex);

lock(&dev->cache.rb_lock);

*** DEADLOCK ***

---
 drivers/infiniband/hw/mlx5/mr.c | 61 +++++++++++++++++++++++++--------
 1 file changed, 47 insertions(+), 14 deletions(-)
drivers/infiniband/hw/mlx5/mr.c +47 -14
index 5fbebafc87742..cb403134eeaea 100644
@@ -2027,23 +2027,50 @@ void mlx5_ib_revoke_data_direct_mrs(struct mlx5_ib_dev *dev)
 	}
 }
 
-static int mlx5_revoke_mr(struct mlx5_ib_mr *mr)
+static int mlx5_umr_revoke_mr_with_lock(struct mlx5_ib_mr *mr)
 {
-	struct mlx5_ib_dev *dev = to_mdev(mr->ibmr.device);
-	struct mlx5_cache_ent *ent = mr->mmkey.cache_ent;
-	bool is_odp = is_odp_mr(mr);
 	bool is_odp_dma_buf = is_dmabuf_mr(mr) &&
-			!to_ib_umem_dmabuf(mr->umem)->pinned;
-	bool from_cache = !!ent;
-	int ret = 0;
+			      !to_ib_umem_dmabuf(mr->umem)->pinned;
+	bool is_odp = is_odp_mr(mr);
+	int ret;
 
 	if (is_odp)
 		mutex_lock(&to_ib_umem_odp(mr->umem)->umem_mutex);
 
 	if (is_odp_dma_buf)
-		dma_resv_lock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv, NULL);
+		dma_resv_lock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv,
+			      NULL);
+
+	ret = mlx5r_umr_revoke_mr(mr);
+
+	if (is_odp) {
+		if (!ret)
+			to_ib_umem_odp(mr->umem)->private = NULL;
+		mutex_unlock(&to_ib_umem_odp(mr->umem)->umem_mutex);
+	}
+
+	if (is_odp_dma_buf) {
+		if (!ret)
+			to_ib_umem_dmabuf(mr->umem)->private = NULL;
+		dma_resv_unlock(
+			to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv);
+	}
 
-	if (mr->mmkey.cacheable && !mlx5r_umr_revoke_mr(mr) && !cache_ent_find_and_store(dev, mr)) {
+	return ret;
+}
+
+static int mlx5r_handle_mkey_cleanup(struct mlx5_ib_mr *mr)
+{
+	bool is_odp_dma_buf = is_dmabuf_mr(mr) &&
+			      !to_ib_umem_dmabuf(mr->umem)->pinned;
+	struct mlx5_ib_dev *dev = to_mdev(mr->ibmr.device);
+	struct mlx5_cache_ent *ent = mr->mmkey.cache_ent;
+	bool is_odp = is_odp_mr(mr);
+	bool from_cache = !!ent;
+	int ret;
+
+	if (mr->mmkey.cacheable && !mlx5_umr_revoke_mr_with_lock(mr) &&
+	    !cache_ent_find_and_store(dev, mr)) {
 		ent = mr->mmkey.cache_ent;
 		/* upon storing to a clean temp entry - schedule its cleanup */
 		spin_lock_irq(&ent->mkeys_queue.lock);
@@ -2055,7 +2082,7 @@ static int mlx5_revoke_mr(struct mlx5_ib_mr *mr)
 			ent->tmp_cleanup_scheduled = true;
 		}
 		spin_unlock_irq(&ent->mkeys_queue.lock);
-		goto out;
+		return 0;
 	}
 
 	if (ent) {
@@ -2064,8 +2091,14 @@ static int mlx5_revoke_mr(struct mlx5_ib_mr *mr)
 		mr->mmkey.cache_ent = NULL;
 		spin_unlock_irq(&ent->mkeys_queue.lock);
 	}
+
+	if (is_odp)
+		mutex_lock(&to_ib_umem_odp(mr->umem)->umem_mutex);
+
+	if (is_odp_dma_buf)
+		dma_resv_lock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv,
+			      NULL);
 	ret = destroy_mkey(dev, mr);
-out:
 	if (is_odp) {
 		if (!ret)
 			to_ib_umem_odp(mr->umem)->private = NULL;
@@ -2075,9 +2108,9 @@ static int mlx5_revoke_mr(struct mlx5_ib_mr *mr)
 	if (is_odp_dma_buf) {
 		if (!ret)
 			to_ib_umem_dmabuf(mr->umem)->private = NULL;
-		dma_resv_unlock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv);
+		dma_resv_unlock(
+			to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv);
 	}
-
 	return ret;
 }
 
@@ -2126,7 +2159,7 @@ static int __mlx5_ib_dereg_mr(struct ib_mr *ibmr)
 	}
 
 	/* Stop DMA */
-	rc = mlx5_revoke_mr(mr);
+	rc = mlx5r_handle_mkey_cleanup(mr);
 	if (rc)
 		return rc;
 
Signature · 1 line
2.39.5
author maintainer
#55 lore raw

PATCH 55/178 RDMA/mlx5: Fix HW counters query for non-representor devices

In reply to Greg Kroah-Hartman

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

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

From: Patrisious Haddad <[email protected]>

[ Upstream commit 3cc1dbfddf88dc5ecce0a75185061403b1f7352d ]

To get the device HW counters, a non-representor switchdev device should use the mlx5_ib_query_q_counters() function and query all of the available counters. While a representor device in switchdev mode should use the mlx5_ib_query_q_counters_vport() function and query only the Q_Counters without the PPCNT counters and congestion control counters, since they aren't relevant for a representor device.

Currently a non-representor switchdev device skips querying the PPCNT counters and congestion control counters, leaving them unupdated. Fix that by properly querying those counters for non-representor devices.

---
 drivers/infiniband/hw/mlx5/counters.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/hw/mlx5/counters.c +1 -1
index b847084dcd998..943e9eb2ad20d 100644
@@ -398,7 +398,7 @@ static int do_get_hw_stats(struct ib_device *ibdev,
 		return ret;
 
 	/* We don't expose device counters over Vports */
-	if (is_mdev_switchdev_mode(dev->mdev) && port_num != 0)
+	if (is_mdev_switchdev_mode(dev->mdev) && dev->is_rep && port_num != 0)
 		goto done;
 
 	if (MLX5_CAP_PCAM_FEATURE(dev->mdev, rx_icrc_encapsulated_counter)) {
Signature · 1 line
2.39.5
author maintainer
#56 lore raw

PATCH 56/178 RDMA/mlx5: Fix CC counters query for MPV

In reply to Greg Kroah-Hartman

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

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

From: Patrisious Haddad <[email protected]>

[ Upstream commit acd245b1e33fc4b9d0f2e3372021d632f7ee0652 ]

In case, CC counters are querying for the second port use the correct core device for the query instead of always using the master core device.

---
 drivers/infiniband/hw/mlx5/counters.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/hw/mlx5/counters.c +1 -1
index 943e9eb2ad20d..a506fafd2b151 100644
@@ -418,7 +418,7 @@ static int do_get_hw_stats(struct ib_device *ibdev,
 			 */
 			goto done;
 		}
-		ret = mlx5_lag_query_cong_counters(dev->mdev,
+		ret = mlx5_lag_query_cong_counters(mdev,
 						   stats->value +
 						   cnts->num_q_counters,
 						   cnts->num_cong_counters,
Signature · 1 line
2.39.5
author maintainer
#39 lore raw

PATCH 39/178 arm64: dts: apple: t8103: Fix PCIe BCM4377 nodename

In reply to Greg Kroah-Hartman

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

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

From: Janne Grunau <[email protected]>

[ Upstream commit ac1daa91e9370e3b88ef7826a73d62a4d09e2717 ]

Fix the following make dtbs_check warnings for all t8103 based devices:

arch/arm64/boot/dts/apple/t8103-j274.dtb: network@0,0: $nodename:0: 'network@0,0' does not match '^wifi(@.*)?$'
from schema $id: http://devicetree.org/schemas/net/wireless/brcm,bcm4329-fmac.yaml#
arch/arm64/boot/dts/apple/t8103-j274.dtb: network@0,0: Unevaluated properties are not allowed ('local-mac-address' was unexpected)
from schema $id: http://devicetree.org/schemas/net/wireless/brcm,bcm4329-fmac.yaml#

Fixes bf2c05b619ff ("arm64: dts: apple: t8103: Expose PCI node for the WiFi MAC address")
---
 arch/arm64/boot/dts/apple/t8103-jxxx.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/apple/t8103-jxxx.dtsi +1 -1
index 8e82231acab59..0c8206156bfef 100644
@@ -71,7 +71,7 @@ hpm1: usb-pd@3f {
  */
 &port00 {
 	bus-range = <1 1>;
-	wifi0: network@0,0 {
+	wifi0: wifi@0,0 {
 		compatible = "pci14e4,4425";
 		reg = <0x10000 0x0 0x0 0x0 0x0>;
 		/* To be filled by the loader */
Signature · 1 line
2.39.5
author maintainer
#57 lore raw

PATCH 57/178 RDMA/mlx5: Fix vport loopback for MPV device

In reply to Greg Kroah-Hartman

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

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

From: Patrisious Haddad <[email protected]>

[ Upstream commit a9a9e68954f29b1e197663f76289db4879fd51bb ]

Always enable vport loopback for both MPV devices on driver start.

Previously in some cases related to MPV RoCE, packets weren't correctly executing loopback check at vport in FW, since it was disabled. Due to complexity of identifying such cases for MPV always enable vport loopback for both GVMIs when binding the slave to the master port.

---
 drivers/infiniband/hw/mlx5/main.c | 33 +++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)
drivers/infiniband/hw/mlx5/main.c +33 -0
index d07cacaa0abd0..4ffe3afb560c9 100644
@@ -1779,6 +1779,33 @@ static void deallocate_uars(struct mlx5_ib_dev *dev,
 					     context->devx_uid);
 }
 
+static int mlx5_ib_enable_lb_mp(struct mlx5_core_dev *master,
+				struct mlx5_core_dev *slave)
+{
+	int err;
+
+	err = mlx5_nic_vport_update_local_lb(master, true);
+	if (err)
+		return err;
+
+	err = mlx5_nic_vport_update_local_lb(slave, true);
+	if (err)
+		goto out;
+
+	return 0;
+
+out:
+	mlx5_nic_vport_update_local_lb(master, false);
+	return err;
+}
+
+static void mlx5_ib_disable_lb_mp(struct mlx5_core_dev *master,
+				  struct mlx5_core_dev *slave)
+{
+	mlx5_nic_vport_update_local_lb(slave, false);
+	mlx5_nic_vport_update_local_lb(master, false);
+}
+
 int mlx5_ib_enable_lb(struct mlx5_ib_dev *dev, bool td, bool qp)
 {
 	int err = 0;
@@ -3483,6 +3510,8 @@ static void mlx5_ib_unbind_slave_port(struct mlx5_ib_dev *ibdev,
 
 	lockdep_assert_held(&mlx5_ib_multiport_mutex);
 
+	mlx5_ib_disable_lb_mp(ibdev->mdev, mpi->mdev);
+
 	mlx5_core_mp_event_replay(ibdev->mdev,
 				  MLX5_DRIVER_EVENT_AFFILIATION_REMOVED,
 				  NULL);
@@ -3578,6 +3607,10 @@ static bool mlx5_ib_bind_slave_port(struct mlx5_ib_dev *ibdev,
 				  MLX5_DRIVER_EVENT_AFFILIATION_DONE,
 				  &key);
 
+	err = mlx5_ib_enable_lb_mp(ibdev->mdev, mpi->mdev);
+	if (err)
+		goto unbind;
+
 	return true;
 
 unbind:
Signature · 1 line
2.39.5
author maintainer
#58 lore raw

PATCH 58/178 platform/mellanox: mlxbf-pmc: Fix duplicate event ID for CACHE_DATA1

In reply to Greg Kroah-Hartman

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

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

From: Alok Tiwari <[email protected]>

[ Upstream commit 173bbec6693f3f3f00dac144f3aa0cd62fb60d33 ]

same ID (103) was assigned to both GDC_BANK0_G_RSE_PIPE_CACHE_DATA0 and GDC_BANK0_G_RSE_PIPE_CACHE_DATA1. This could lead to incorrect event mapping.
Updated the ID to 104 to ensure uniqueness.

Fixes 423c3361855c ("platform/mellanox: mlxbf-pmc: Add support for BlueField-3")
---
 drivers/platform/mellanox/mlxbf-pmc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/mellanox/mlxbf-pmc.c +1 -1
index 36a00692347dc..771a24c397c12 100644
@@ -713,7 +713,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_llt_events[] = {
 	{101, "GDC_BANK0_HIT_DCL_PARTIAL"},
 	{102, "GDC_BANK0_EVICT_DCL"},
 	{103, "GDC_BANK0_G_RSE_PIPE_CACHE_DATA0"},
-	{103, "GDC_BANK0_G_RSE_PIPE_CACHE_DATA1"},
+	{104, "GDC_BANK0_G_RSE_PIPE_CACHE_DATA1"},
 	{105, "GDC_BANK0_ARB_STRB"},
 	{106, "GDC_BANK0_ARB_WAIT"},
 	{107, "GDC_BANK0_GGA_STRB"},
Signature · 1 line
2.39.5
author maintainer
#59 lore raw

PATCH 59/178 platform/mellanox: nvsw-sn2201: Fix bus number in adapter error message

In reply to Greg Kroah-Hartman

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

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

From: Alok Tiwari <[email protected]>

[ Upstream commit d07143b507c51c04c091081627c5a130e9d3c517 ]

change error log to use correct bus number from main_mux_devs instead of cpld_devs.

Fixes 662f24826f95 ("platform/mellanox: Add support for new SN2201 system")
---
 drivers/platform/mellanox/nvsw-sn2201.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/mellanox/nvsw-sn2201.c +1 -1
index 0c047aa2345b3..490001a9afd76 100644
@@ -1088,7 +1088,7 @@ static int nvsw_sn2201_i2c_completion_notify(void *handle, int id)
 	if (!nvsw_sn2201->main_mux_devs->adapter) {
 		err = -ENODEV;
 		dev_err(nvsw_sn2201->dev, "Failed to get adapter for bus %d\n",
-			nvsw_sn2201->cpld_devs->nr);
+			nvsw_sn2201->main_mux_devs->nr);
 		goto i2c_get_adapter_main_fail;
 	}
 
Signature · 1 line
2.39.5
author maintainer
#60 lore raw

PATCH 60/178 platform/x86: wmi: Fix WMI event enablement

In reply to Greg Kroah-Hartman

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

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

From: Armin Wolf <[email protected]>

[ Upstream commit cf0b812500e64a7d5e2957abed38c3a97917b34f ]

It turns out that the Windows WMI-ACPI driver always enables/disables WMI events regardless of whether they are marked as expensive or not. This finding is further reinforced when reading the documentation of the WMI_FUNCTION_CONTROL_CALLBACK callback used by Windows drivers for enabling/disabling WMI devices:

The DpWmiFunctionControl routine enables or disables notification of events, and enables or disables data collection for data blocks that the driver registered as expensive to collect.

Follow this behavior to fix the WMI event used for reporting hotkey events on the Dell Latitude 5400 and likely many more devices.

---
 drivers/platform/x86/wmi.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)
drivers/platform/x86/wmi.c +11 -5
index e46453750d5f1..03aecf8bb7f8e 100644
@@ -177,16 +177,22 @@ static int wmi_device_enable(struct wmi_device *wdev, bool enable)
 	acpi_handle handle;
 	acpi_status status;
 
-	if (!(wblock->gblock.flags & ACPI_WMI_EXPENSIVE))
-		return 0;
-
 	if (wblock->dev.dev.type == &wmi_type_method)
 		return 0;
 
-	if (wblock->dev.dev.type == &wmi_type_event)
+	if (wblock->dev.dev.type == &wmi_type_event) {
+		/*
+		 * Windows always enables/disables WMI events, even when they are
+		 * not marked as being expensive. We follow this behavior for
+		 * compatibility reasons.
+		 */
 		snprintf(method, sizeof(method), "WE%02X", wblock->gblock.notify_id);
-	else
+	} else {
+		if (!(wblock->gblock.flags & ACPI_WMI_EXPENSIVE))
+			return 0;
+
 		get_acpi_method_name(wblock, 'C', method);
+	}
 
 	/*
 	 * Not all WMI devices marked as expensive actually implement the
Signature · 1 line
2.39.5
author maintainer
#61 lore raw

PATCH 61/178 flexfiles/pNFS: update stats on NFS4ERR_DELAY for v4.1 DSes

In reply to Greg Kroah-Hartman

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

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

From: Tigran Mkrtchyan <[email protected]>

[ Upstream commit e3e3775392f3f0f3e3044f8c162bf47858e01759 ]

On NFS4ERR_DELAY nfs slient updates its stats, but misses for flexfiles v4.1 DSes.

Stable-dep-of: 38074de35b01 ("NFSv4/flexfiles: Fix handling of NFS level errors in I/O")

---
 fs/nfs/flexfilelayout/flexfilelayout.c | 2 ++
 1 file changed, 2 insertions(+)
fs/nfs/flexfilelayout/flexfilelayout.c +2 -0
index e6909cafab686..df48074605967 100644
@@ -1129,6 +1129,8 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task,
 		nfs4_schedule_session_recovery(clp->cl_session, task->tk_status);
 		break;
 	case -NFS4ERR_DELAY:
+		nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY);
+		fallthrough;
 	case -NFS4ERR_GRACE:
 		rpc_delay(task, FF_LAYOUT_POLL_RETRY_MAX);
 		break;
Signature · 1 line
2.39.5
author maintainer
#62 lore raw

PATCH 62/178 NFSv4/flexfiles: Fix handling of NFS level errors in I/O

In reply to Greg Kroah-Hartman

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

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

From: Trond Myklebust <[email protected]>

[ Upstream commit 38074de35b015df5623f524d6f2b49a0cd395c40 ]

Allow the flexfiles error handling to recognise NFS level errors (as opposed to RPC level errors) and handle them separately. The main motivator is the NFSERR_PERM errors that get returned if the NFS client connects to the data server through a port number that is lower than 1024. In that case, the client should disconnect and retry a READ on a different data server, or it should retry a WRITE after reconnecting.

Fixes d67ae825a59d ("pnfs/flexfiles: Add the FlexFile Layout Driver")
---
 fs/nfs/flexfilelayout/flexfilelayout.c | 118 ++++++++++++++++++-------
 1 file changed, 84 insertions(+), 34 deletions(-)
fs/nfs/flexfilelayout/flexfilelayout.c +84 -34
index df48074605967..4bea008dbebd7 100644
@@ -1105,6 +1105,7 @@ static void ff_layout_reset_read(struct nfs_pgio_header *hdr)
 }
 
 static int ff_layout_async_handle_error_v4(struct rpc_task *task,
+					   u32 op_status,
 					   struct nfs4_state *state,
 					   struct nfs_client *clp,
 					   struct pnfs_layout_segment *lseg,
@@ -1115,34 +1116,42 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task,
 	struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx);
 	struct nfs4_slot_table *tbl = &clp->cl_session->fc_slot_table;
 
-	switch (task->tk_status) {
-	case -NFS4ERR_BADSESSION:
-	case -NFS4ERR_BADSLOT:
-	case -NFS4ERR_BAD_HIGH_SLOT:
-	case -NFS4ERR_DEADSESSION:
-	case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
-	case -NFS4ERR_SEQ_FALSE_RETRY:
-	case -NFS4ERR_SEQ_MISORDERED:
+	switch (op_status) {
+	case NFS4_OK:
+	case NFS4ERR_NXIO:
+		break;
+	case NFSERR_PERM:
+		if (!task->tk_xprt)
+			break;
+		xprt_force_disconnect(task->tk_xprt);
+		goto out_retry;
+	case NFS4ERR_BADSESSION:
+	case NFS4ERR_BADSLOT:
+	case NFS4ERR_BAD_HIGH_SLOT:
+	case NFS4ERR_DEADSESSION:
+	case NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+	case NFS4ERR_SEQ_FALSE_RETRY:
+	case NFS4ERR_SEQ_MISORDERED:
 		dprintk("%s ERROR %d, Reset session. Exchangeid "
 			"flags 0x%x\n", __func__, task->tk_status,
 			clp->cl_exchange_flags);
 		nfs4_schedule_session_recovery(clp->cl_session, task->tk_status);
-		break;
-	case -NFS4ERR_DELAY:
+		goto out_retry;
+	case NFS4ERR_DELAY:
 		nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY);
 		fallthrough;
-	case -NFS4ERR_GRACE:
+	case NFS4ERR_GRACE:
 		rpc_delay(task, FF_LAYOUT_POLL_RETRY_MAX);
-		break;
-	case -NFS4ERR_RETRY_UNCACHED_REP:
-		break;
+		goto out_retry;
+	case NFS4ERR_RETRY_UNCACHED_REP:
+		goto out_retry;
 	/* Invalidate Layout errors */
-	case -NFS4ERR_PNFS_NO_LAYOUT:
-	case -ESTALE:           /* mapped NFS4ERR_STALE */
-	case -EBADHANDLE:       /* mapped NFS4ERR_BADHANDLE */
-	case -EISDIR:           /* mapped NFS4ERR_ISDIR */
-	case -NFS4ERR_FHEXPIRED:
-	case -NFS4ERR_WRONG_TYPE:
+	case NFS4ERR_PNFS_NO_LAYOUT:
+	case NFS4ERR_STALE:
+	case NFS4ERR_BADHANDLE:
+	case NFS4ERR_ISDIR:
+	case NFS4ERR_FHEXPIRED:
+	case NFS4ERR_WRONG_TYPE:
 		dprintk("%s Invalid layout error %d\n", __func__,
 			task->tk_status);
 		/*
@@ -1155,6 +1164,11 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task,
 		pnfs_destroy_layout(NFS_I(inode));
 		rpc_wake_up(&tbl->slot_tbl_waitq);
 		goto reset;
+	default:
+		break;
+	}
+
+	switch (task->tk_status) {
 	/* RPC connection errors */
 	case -ENETDOWN:
 	case -ENETUNREACH:
@@ -1174,27 +1188,56 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task,
 		nfs4_delete_deviceid(devid->ld, devid->nfs_client,
 				&devid->deviceid);
 		rpc_wake_up(&tbl->slot_tbl_waitq);
-		fallthrough;
+		break;
 	default:
-		if (ff_layout_avoid_mds_available_ds(lseg))
-			return -NFS4ERR_RESET_TO_PNFS;
-reset:
-		dprintk("%s Retry through MDS. Error %d\n", __func__,
-			task->tk_status);
-		return -NFS4ERR_RESET_TO_MDS;
+		break;
 	}
+
+	if (ff_layout_avoid_mds_available_ds(lseg))
+		return -NFS4ERR_RESET_TO_PNFS;
+reset:
+	dprintk("%s Retry through MDS. Error %d\n", __func__,
+		task->tk_status);
+	return -NFS4ERR_RESET_TO_MDS;
+
+out_retry:
 	task->tk_status = 0;
 	return -EAGAIN;
 }
 
 /* Retry all errors through either pNFS or MDS except for -EJUKEBOX */
 static int ff_layout_async_handle_error_v3(struct rpc_task *task,
+					   u32 op_status,
 					   struct nfs_client *clp,
 					   struct pnfs_layout_segment *lseg,
 					   u32 idx)
 {
 	struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx);
 
+	switch (op_status) {
+	case NFS_OK:
+	case NFSERR_NXIO:
+		break;
+	case NFSERR_PERM:
+		if (!task->tk_xprt)
+			break;
+		xprt_force_disconnect(task->tk_xprt);
+		goto out_retry;
+	case NFSERR_ACCES:
+	case NFSERR_BADHANDLE:
+	case NFSERR_FBIG:
+	case NFSERR_IO:
+	case NFSERR_NOSPC:
+	case NFSERR_ROFS:
+	case NFSERR_STALE:
+		goto out_reset_to_pnfs;
+	case NFSERR_JUKEBOX:
+		nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY);
+		goto out_retry;
+	default:
+		break;
+	}
+
 	switch (task->tk_status) {
 	/* File access problems. Don't mark the device as unavailable */
 	case -EACCES:
@@ -1218,6 +1261,7 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task,
 		nfs4_delete_deviceid(devid->ld, devid->nfs_client,
 				&devid->deviceid);
 	}
+out_reset_to_pnfs:
 	/* FIXME: Need to prevent infinite looping here. */
 	return -NFS4ERR_RESET_TO_PNFS;
 out_retry:
@@ -1228,6 +1272,7 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task,
 }
 
 static int ff_layout_async_handle_error(struct rpc_task *task,
+					u32 op_status,
 					struct nfs4_state *state,
 					struct nfs_client *clp,
 					struct pnfs_layout_segment *lseg,
@@ -1246,10 +1291,11 @@ static int ff_layout_async_handle_error(struct rpc_task *task,
 
 	switch (vers) {
 	case 3:
-		return ff_layout_async_handle_error_v3(task, clp, lseg, idx);
-	case 4:
-		return ff_layout_async_handle_error_v4(task, state, clp,
+		return ff_layout_async_handle_error_v3(task, op_status, clp,
 						       lseg, idx);
+	case 4:
+		return ff_layout_async_handle_error_v4(task, op_status, state,
+						       clp, lseg, idx);
 	default:
 		/* should never happen */
 		WARN_ON_ONCE(1);
@@ -1302,6 +1348,7 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 	switch (status) {
 	case NFS4ERR_DELAY:
 	case NFS4ERR_GRACE:
+	case NFS4ERR_PERM:
 		break;
 	case NFS4ERR_NXIO:
 		ff_layout_mark_ds_unreachable(lseg, idx);
@@ -1334,7 +1381,8 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 		trace_ff_layout_read_error(hdr, task->tk_status);
 	}
 
-	err = ff_layout_async_handle_error(task, hdr->args.context->state,
+	err = ff_layout_async_handle_error(task, hdr->res.op_status,
+					   hdr->args.context->state,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
 
@@ -1507,7 +1555,8 @@ static int ff_layout_write_done_cb(struct rpc_task *task,
 		trace_ff_layout_write_error(hdr, task->tk_status);
 	}
 
-	err = ff_layout_async_handle_error(task, hdr->args.context->state,
+	err = ff_layout_async_handle_error(task, hdr->res.op_status,
+					   hdr->args.context->state,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
 
@@ -1556,8 +1605,9 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 		trace_ff_layout_commit_error(data, task->tk_status);
 	}
 
-	err = ff_layout_async_handle_error(task, NULL, data->ds_clp,
-					   data->lseg, data->ds_commit_index);
+	err = ff_layout_async_handle_error(task, data->res.op_status,
+					   NULL, data->ds_clp, data->lseg,
+					   data->ds_commit_index);
 
 	trace_nfs4_pnfs_commit_ds(data, err);
 	switch (err) {
Signature · 1 line
2.39.5
author maintainer
#63 lore raw

PATCH 63/178 Bluetooth: Prevent unintended pause by checking if advertising is active

In reply to Greg Kroah-Hartman

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

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

From: Yang Li <[email protected]>

[ Upstream commit 1f029b4e30a602db33dedee5ac676e9236ad193c ]

When PA Create Sync is enabled, advertising resumes unexpectedly. Therefore, it's necessary to check whether advertising is currently active before attempting to pause it.

< HCI Command: LE Add Device To... (0x08|0x0011) plen 7 #1345 [hci0] 48.306205
Address type: Random (0x01)
Address: 4F:84:84:5F:88:17 (Resolvable)
Identity type: Random (0x01)
Identity: FC:5B:8C:F7:5D:FB (Static)
< HCI Command: LE Set Address Re.. (0x08|0x002d) plen 1 #1347 [hci0] 48.308023
Address resolution: Enabled (0x01)
...
< HCI Command: LE Set Extended A.. (0x08|0x0039) plen 6 #1349 [hci0] 48.309650
Extended advertising: Enabled (0x01)
Number of sets: 1 (0x01)
Entry 0
Handle: 0x01
Duration: 0 ms (0x00)
Max ext adv events: 0
...
< HCI Command: LE Periodic Adve.. (0x08|0x0044) plen 14 #1355 [hci0] 48.314575
Options: 0x0000
Use advertising SID, Advertiser Address Type and address Reporting initially enabled
SID: 0x02
Adv address type: Random (0x01)
Adv address: 4F:84:84:5F:88:17 (Resolvable)
Identity type: Random (0x01)
Identity: FC:5B:8C:F7:5D:FB (Static)
Skip: 0x0000
Sync timeout: 20000 msec (0x07d0)
Sync CTE type: 0x0000

Fixes ad383c2c65a5 ("Bluetooth: hci_sync: Enable advertising when LL privacy is enabled")
---
 net/bluetooth/hci_sync.c | 4 ++++
 1 file changed, 4 insertions(+)
net/bluetooth/hci_sync.c +4 -0
index 3660da3b241a7..9955d6cd7b76f 100644
@@ -2507,6 +2507,10 @@ static int hci_pause_advertising_sync(struct hci_dev *hdev)
 	int err;
 	int old_state;
 
+	/* If controller is not advertising we are done. */
+	if (!hci_dev_test_flag(hdev, HCI_LE_ADV))
+		return 0;
+
 	/* If already been paused there is nothing to do. */
 	if (hdev->advertising_paused)
 		return 0;
Signature · 1 line
2.39.5
author maintainer
#64 lore raw

PATCH 64/178 btrfs: fix failure to rebuild free space tree using multiple transactions

In reply to Greg Kroah-Hartman

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

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

From: Filipe Manana <[email protected]>

[ Upstream commit 1e6ed33cabba8f06f532f2e5851a102602823734 ]

If we are rebuilding a free space tree, while modifying the free space tree we may need to allocate a new metadata block group.
If we end up using multiple transactions for the rebuild, when we call btrfs_end_transaction() we enter btrfs_create_pending_block_groups() which calls add_block_group_free_space() to add items to the free space tree for the block group.

Then later during the free space tree rebuild, at btrfs_rebuild_free_space_tree(), we may find such new block groups and call populate_free_space_tree() for them, which fails with -EEXIST because there are already items in the free space tree. Then we abort the transaction with -EEXIST at btrfs_rebuild_free_space_tree(). Notice that we say "may find" the new block groups because a new block group may be inserted in the block groups rbtree, which is being iterated by the rebuild process, before or after the current node where the rebuild process is currently at.

Syzbot recently reported such case which produces a trace like the following:

  ------------[ cut here ]------------
  BTRFS: Transaction aborted (error -17)
  WARNING: CPU: 1 PID: 7626 at fs/btrfs/free-space-tree.c:1341 btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341
  Modules linked in:
  CPU: 1 UID: 0 PID: 7626 Comm: syz.2.25 Not tainted 6.15.0-rc7-syzkaller-00085-gd7fa1af5b33e-dirty #0 PREEMPT
  Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
  pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
  pc : btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341
  lr : btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341
  sp : ffff80009c4f7740
  x29: ffff80009c4f77b0 x28: ffff0000d4c3f400 x27: 0000000000000000
  x26: dfff800000000000 x25: ffff70001389eee8 x24: 0000000000000003
  x23: 1fffe000182b6e7b x22: 0000000000000000 x21: ffff0000c15b73d8
  x20: 00000000ffffffef x19: ffff0000c15b7378 x18: 1fffe0003386f276
  x17: ffff80008f31e000 x16: ffff80008adbe98c x15: 0000000000000001
  x14: 1fffe0001b281550 x13: 0000000000000000 x12: 0000000000000000
  x11: ffff60001b281551 x10: 0000000000000003 x9 : 1c8922000a902c00
  x8 : 1c8922000a902c00 x7 : ffff800080485878 x6 : 0000000000000000
  x5 : 0000000000000001 x4 : 0000000000000001 x3 : ffff80008047843c
  x2 : 0000000000000001 x1 : ffff80008b3ebc40 x0 : 0000000000000001
  Call trace:
   btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341 (P)
   btrfs_start_pre_rw_mount+0xa78/0xe10 fs/btrfs/disk-io.c:3074
   btrfs_remount_rw fs/btrfs/super.c:1319 [inline]
   btrfs_reconfigure+0x828/0x2418 fs/btrfs/super.c:1543
   reconfigure_super+0x1d4/0x6f0 fs/super.c:1083
   do_remount fs/namespace.c:3365 [inline]
   path_mount+0xb34/0xde0 fs/namespace.c:4200
   do_mount fs/namespace.c:4221 [inline]
   __do_sys_mount fs/namespace.c:4432 [inline]
   __se_sys_mount fs/namespace.c:4409 [inline]
   __arm64_sys_mount+0x3e8/0x468 fs/namespace.c:4409
   __invoke_syscall arch/arm64/kernel/syscall.c:35 [inline]
   invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:49
   el0_svc_common+0x130/0x23c arch/arm64/kernel/syscall.c:132
   do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:151
   el0_svc+0x58/0x17c arch/arm64/kernel/entry-common.c:767
   el0t_64_sync_handler+0x78/0x108 arch/arm64/kernel/entry-common.c:786
   el0t_64_sync+0x198/0x19c arch/arm64/kernel/entry.S:600
  irq event stamp: 330
  hardirqs last  enabled at (329): [<ffff80008048590c>] raw_spin_rq_unlock_irq kernel/sched/sched.h:1525 [inline]
  hardirqs last  enabled at (329): [<ffff80008048590c>] finish_lock_switch+0xb0/0x1c0 kernel/sched/core.c:5130
  hardirqs last disabled at (330): [<ffff80008adb9e60>] el1_dbg+0x24/0x80 arch/arm64/kernel/entry-common.c:511
  softirqs last  enabled at (10): [<ffff8000801fbf10>] local_bh_enable+0x10/0x34 include/linux/bottom_half.h:32
  softirqs last disabled at (8): [<ffff8000801fbedc>] local_bh_disable+0x10/0x34 include/linux/bottom_half.h:19
  ---[ end trace 0000000000000000 ]---

Fix this by flagging new block groups which had their free space tree entries already added and then skip them in the rebuild process. Also, since the rebuild may be triggered when doing a remount, make sure that when we clear an existing free space tree that we clear such flag from every existing block group, otherwise we would skip those block groups during the rebuild.

Reported-by [email protected]
Fixes 882af9f13e83 ("btrfs: handle free space tree rebuild in multiple transactions")
---
 fs/btrfs/block-group.h     |  2 ++
 fs/btrfs/free-space-tree.c | 40 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 42 insertions(+)
fs/btrfs/block-group.h +2 -0
index 36937eeab9b8a..08548b0b5ea1d 100644
@@ -83,6 +83,8 @@ enum btrfs_block_group_flags {
 	BLOCK_GROUP_FLAG_ZONED_DATA_RELOC,
 	/* Does the block group need to be added to the free space tree? */
 	BLOCK_GROUP_FLAG_NEEDS_FREE_SPACE,
+	/* Set after we add a new block group to the free space tree. */
+	BLOCK_GROUP_FLAG_FREE_SPACE_ADDED,
 	/* Indicate that the block group is placed on a sequential zone */
 	BLOCK_GROUP_FLAG_SEQUENTIAL_ZONE,
 	/*
fs/btrfs/free-space-tree.c +40 -0
index 39c6b96a4c25a..b65a20fd519ba 100644
@@ -1217,6 +1217,7 @@ static int clear_free_space_tree(struct btrfs_trans_handle *trans,
 {
 	BTRFS_PATH_AUTO_FREE(path);
 	struct btrfs_key key;
+	struct rb_node *node;
 	int nr;
 	int ret;
 
@@ -1245,6 +1246,16 @@ static int clear_free_space_tree(struct btrfs_trans_handle *trans,
 		btrfs_release_path(path);
 	}
 
+	node = rb_first_cached(&trans->fs_info->block_group_cache_tree);
+	while (node) {
+		struct btrfs_block_group *bg;
+
+		bg = rb_entry(node, struct btrfs_block_group, cache_node);
+		clear_bit(BLOCK_GROUP_FLAG_FREE_SPACE_ADDED, &bg->runtime_flags);
+		node = rb_next(node);
+		cond_resched();
+	}
+
 	return 0;
 }
 
@@ -1334,12 +1345,18 @@ int btrfs_rebuild_free_space_tree(struct btrfs_fs_info *fs_info)
 
 		block_group = rb_entry(node, struct btrfs_block_group,
 				       cache_node);
+
+		if (test_bit(BLOCK_GROUP_FLAG_FREE_SPACE_ADDED,
+			     &block_group->runtime_flags))
+			goto next;
+
 		ret = populate_free_space_tree(trans, block_group);
 		if (ret) {
 			btrfs_abort_transaction(trans, ret);
 			btrfs_end_transaction(trans);
 			return ret;
 		}
+next:
 		if (btrfs_should_end_transaction(trans)) {
 			btrfs_end_transaction(trans);
 			trans = btrfs_start_transaction(free_space_root, 1);
@@ -1366,6 +1383,29 @@ static int __add_block_group_free_space(struct btrfs_trans_handle *trans,
 
 	clear_bit(BLOCK_GROUP_FLAG_NEEDS_FREE_SPACE, &block_group->runtime_flags);
 
+	/*
+	 * While rebuilding the free space tree we may allocate new metadata
+	 * block groups while modifying the free space tree.
+	 *
+	 * Because during the rebuild (at btrfs_rebuild_free_space_tree()) we
+	 * can use multiple transactions, every time btrfs_end_transaction() is
+	 * called at btrfs_rebuild_free_space_tree() we finish the creation of
+	 * new block groups by calling btrfs_create_pending_block_groups(), and
+	 * that in turn calls us, through add_block_group_free_space(), to add
+	 * a free space info item and a free space extent item for the block
+	 * group.
+	 *
+	 * Then later btrfs_rebuild_free_space_tree() may find such new block
+	 * groups and processes them with populate_free_space_tree(), which can
+	 * fail with EEXIST since there are already items for the block group in
+	 * the free space tree. Notice that we say "may find" because a new
+	 * block group may be added to the block groups rbtree in a node before
+	 * or after the block group currently being processed by the rebuild
+	 * process. So signal the rebuild process to skip such new block groups
+	 * if it finds them.
+	 */
+	set_bit(BLOCK_GROUP_FLAG_FREE_SPACE_ADDED, &block_group->runtime_flags);
+
 	ret = add_new_free_space_info(trans, block_group, path);
 	if (ret)
 		return ret;
Signature · 1 line
2.39.5
author maintainer
#65 lore raw

PATCH 65/178 btrfs: fix missing error handling when searching for inode refs during log replay

In reply to Greg Kroah-Hartman

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

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

From: Filipe Manana <[email protected]>

[ Upstream commit 6561a40ceced9082f50c374a22d5966cf9fc5f5c ]

During log replay, at __add_inode_ref(), when we are searching for inode ref keys we totally ignore if btrfs_search_slot() returns an error. This may make a log replay succeed when there was an actual error and leave some metadata inconsistency in a subvolume tree. Fix this by checking if an error was returned from btrfs_search_slot() and if so, return it to the caller.

Fixes e02119d5a7b4 ("Btrfs: Add a write ahead tree log to optimize synchronous operations")
---
 fs/btrfs/tree-log.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
fs/btrfs/tree-log.c +3 -1
index ef9660eabf0c6..cc223c3b39c10 100644
@@ -1070,7 +1070,9 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans,
 	search_key.type = BTRFS_INODE_REF_KEY;
 	search_key.offset = parent_objectid;
 	ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0);
-	if (ret == 0) {
+	if (ret < 0) {
+		return ret;
+	} else if (ret == 0) {
 		struct btrfs_inode_ref *victim_ref;
 		unsigned long ptr;
 		unsigned long ptr_end;
Signature · 1 line
2.39.5
author maintainer
#66 lore raw

PATCH 66/178 btrfs: fix iteration of extrefs during log replay

In reply to Greg Kroah-Hartman

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

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

From: Filipe Manana <[email protected]>

[ Upstream commit 54a7081ed168b72a8a2d6ef4ba3a1259705a2926 ]

At __inode_add_ref() when processing extrefs, if we jump into the next label we have an undefined value of victim_name.len, since we haven't initialized it before we did the goto. This results in an invalid memory access in the next iteration of the loop since victim_name.len was not initialized to the length of the name of the current extref.

Fix this by initializing victim_name.len with the current extref's name length.

Fixes e43eec81c516 ("btrfs: use struct qstr instead of name and namelen pairs")
---
 fs/btrfs/tree-log.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/btrfs/tree-log.c +2 -2
index cc223c3b39c10..118db10464207 100644
@@ -1145,13 +1145,13 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans,
 			struct fscrypt_str victim_name;
 
 			extref = (struct btrfs_inode_extref *)(base + cur_offset);
+			victim_name.len = btrfs_inode_extref_name_len(leaf, extref);
 
 			if (btrfs_inode_extref_parent(leaf, extref) != parent_objectid)
 				goto next;
 
 			ret = read_alloc_one_name(leaf, &extref->name,
-				 btrfs_inode_extref_name_len(leaf, extref),
-				 &victim_name);
+						  victim_name.len, &victim_name);
 			if (ret)
 				return ret;
 
Signature · 1 line
2.39.5
author maintainer
#40 lore raw

PATCH 40/178 arm64: dts: apple: Drop {address,size}-cells from SPI NOR

In reply to Greg Kroah-Hartman

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

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

From: Sven Peter <[email protected]>

[ Upstream commit 811a909978bf59caa25359e0aca4e30500dcff26 ]

Fix the following warning by dropping #{address,size}-cells from the SPI NOR node which only has a single child node without reg property:

spi1-nvram.dtsi:19.10-38.4: Warning (avoid_unnecessary_addr_size): /soc/spi@235104000/flash@0: unnecessary #address-cells/#size-cells without "ranges", "dma-ranges" or child "reg" property

Fixes 3febe9de5ca5 ("arm64: dts: apple: Add SPI NOR nvram partition to all devices")
---
 arch/arm64/boot/dts/apple/spi1-nvram.dtsi | 2 --
 1 file changed, 2 deletions(-)
arch/arm64/boot/dts/apple/spi1-nvram.dtsi +0 -2
index 3df2fd3993b52..9740fbf200f0b 100644
@@ -20,8 +20,6 @@ flash@0 {
 		compatible = "jedec,spi-nor";
 		reg = <0x0>;
 		spi-max-frequency = <25000000>;
-		#address-cells = <1>;
-		#size-cells = <1>;
 
 		partitions {
 			compatible = "fixed-partitions";
Signature · 1 line
2.39.5
author maintainer
#67 lore raw

PATCH 67/178 btrfs: fix inode lookup error handling during log replay

In reply to Greg Kroah-Hartman

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

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

From: Filipe Manana <[email protected]>

[ Upstream commit 5f61b961599acbd2bed028d3089105a1f7d224b8 ]

When replaying log trees we use read_one_inode() to get an inode, which is just a wrapper around btrfs_iget_logging(), which in turn is a wrapper for btrfs_iget(). But read_one_inode() always returns NULL for any error that btrfs_iget_logging() / btrfs_iget() may return and this is a problem because:

1) In many callers of read_one_inode() we convert the NULL into -EIO, which is not accurate since btrfs_iget() may return -ENOMEM and -ENOENT for example, besides -EIO and other errors. So during log replay we may end up reporting a false -EIO, which is confusing since we may not have had any IO error at all;

2) When replaying directory deletes, at replay_dir_deletes(), we assume the NULL returned from read_one_inode() means that the inode doesn't exist and then proceed as if no error had happened. This is wrong because unless btrfs_iget() returned ERR_PTR(-ENOENT), we had an actual error and the target inode may exist in the target subvolume root - this may later result in the log replay code failing at a later stage (if we are "lucky") or succeed but leaving some inconsistency in the filesystem.

So fix this by not ignoring errors from btrfs_iget_logging() and as a consequence remove the read_one_inode() wrapper and just use btrfs_iget_logging() directly. Also since btrfs_iget_logging() is supposed to be called only against subvolume roots, just like read_one_inode() which had a comment about it, add an assertion to btrfs_iget_logging() to check that the target root corresponds to a subvolume root.

Fixes 5d4f98a28c7d ("Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE)")
---
 fs/btrfs/tree-log.c | 127 +++++++++++++++++++++-----------------------
 1 file changed, 62 insertions(+), 65 deletions(-)
fs/btrfs/tree-log.c +62 -65
index 118db10464207..f8611ec11da9d 100644
@@ -143,6 +143,9 @@ static struct btrfs_inode *btrfs_iget_logging(u64 objectid, struct btrfs_root *r
 	unsigned int nofs_flag;
 	struct btrfs_inode *inode;
 
+	/* Only meant to be called for subvolume roots and not for log roots. */
+	ASSERT(is_fstree(btrfs_root_id(root)));
+
 	/*
 	 * We're holding a transaction handle whether we are logging or
 	 * replaying a log tree, so we must make sure NOFS semantics apply
@@ -604,21 +607,6 @@ static int read_alloc_one_name(struct extent_buffer *eb, void *start, int len,
 	return 0;
 }
 
-/*
- * simple helper to read an inode off the disk from a given root
- * This can only be called for subvolume roots and not for the log
- */
-static noinline struct btrfs_inode *read_one_inode(struct btrfs_root *root,
-						   u64 objectid)
-{
-	struct btrfs_inode *inode;
-
-	inode = btrfs_iget_logging(objectid, root);
-	if (IS_ERR(inode))
-		return NULL;
-	return inode;
-}
-
 /* replays a single extent in 'eb' at 'slot' with 'key' into the
  * subvolume 'root'.  path is released on entry and should be released
  * on exit.
@@ -671,9 +659,9 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 		return 0;
 	}
 
-	inode = read_one_inode(root, key->objectid);
-	if (!inode)
-		return -EIO;
+	inode = btrfs_iget_logging(key->objectid, root);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
 
 	/*
 	 * first check to see if we already have this extent in the
@@ -945,9 +933,10 @@ static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans,
 
 	btrfs_release_path(path);
 
-	inode = read_one_inode(root, location.objectid);
-	if (!inode) {
-		ret = -EIO;
+	inode = btrfs_iget_logging(location.objectid, root);
+	if (IS_ERR(inode)) {
+		ret = PTR_ERR(inode);
+		inode = NULL;
 		goto out;
 	}
 
@@ -1166,10 +1155,10 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans,
 				kfree(victim_name.name);
 				return ret;
 			} else if (!ret) {
-				ret = -ENOENT;
-				victim_parent = read_one_inode(root,
-						parent_objectid);
-				if (victim_parent) {
+				victim_parent = btrfs_iget_logging(parent_objectid, root);
+				if (IS_ERR(victim_parent)) {
+					ret = PTR_ERR(victim_parent);
+				} else {
 					inc_nlink(&inode->vfs_inode);
 					btrfs_release_path(path);
 
@@ -1314,9 +1303,9 @@ static int unlink_old_inode_refs(struct btrfs_trans_handle *trans,
 			struct btrfs_inode *dir;
 
 			btrfs_release_path(path);
-			dir = read_one_inode(root, parent_id);
-			if (!dir) {
-				ret = -ENOENT;
+			dir = btrfs_iget_logging(parent_id, root);
+			if (IS_ERR(dir)) {
+				ret = PTR_ERR(dir);
 				kfree(name.name);
 				goto out;
 			}
@@ -1388,15 +1377,17 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 	 * copy the back ref in.  The link count fixup code will take
 	 * care of the rest
 	 */
-	dir = read_one_inode(root, parent_objectid);
-	if (!dir) {
-		ret = -ENOENT;
+	dir = btrfs_iget_logging(parent_objectid, root);
+	if (IS_ERR(dir)) {
+		ret = PTR_ERR(dir);
+		dir = NULL;
 		goto out;
 	}
 
-	inode = read_one_inode(root, inode_objectid);
-	if (!inode) {
-		ret = -EIO;
+	inode = btrfs_iget_logging(inode_objectid, root);
+	if (IS_ERR(inode)) {
+		ret = PTR_ERR(inode);
+		inode = NULL;
 		goto out;
 	}
 
@@ -1408,11 +1399,13 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 			 * parent object can change from one array
 			 * item to another.
 			 */
-			if (!dir)
-				dir = read_one_inode(root, parent_objectid);
 			if (!dir) {
-				ret = -ENOENT;
-				goto out;
+				dir = btrfs_iget_logging(parent_objectid, root);
+				if (IS_ERR(dir)) {
+					ret = PTR_ERR(dir);
+					dir = NULL;
+					goto out;
+				}
 			}
 		} else {
 			ret = ref_get_fields(eb, ref_ptr, &name, &ref_index);
@@ -1681,9 +1674,9 @@ static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans,
 			break;
 
 		btrfs_release_path(path);
-		inode = read_one_inode(root, key.offset);
-		if (!inode) {
-			ret = -EIO;
+		inode = btrfs_iget_logging(key.offset, root);
+		if (IS_ERR(inode)) {
+			ret = PTR_ERR(inode);
 			break;
 		}
 
@@ -1719,9 +1712,9 @@ static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans,
 	struct btrfs_inode *inode;
 	struct inode *vfs_inode;
 
-	inode = read_one_inode(root, objectid);
-	if (!inode)
-		return -EIO;
+	inode = btrfs_iget_logging(objectid, root);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
 
 	vfs_inode = &inode->vfs_inode;
 	key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID;
@@ -1760,14 +1753,14 @@ static noinline int insert_one_name(struct btrfs_trans_handle *trans,
 	struct btrfs_inode *dir;
 	int ret;
 
-	inode = read_one_inode(root, location->objectid);
-	if (!inode)
-		return -ENOENT;
+	inode = btrfs_iget_logging(location->objectid, root);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
 
-	dir = read_one_inode(root, dirid);
-	if (!dir) {
+	dir = btrfs_iget_logging(dirid, root);
+	if (IS_ERR(dir)) {
 		iput(&inode->vfs_inode);
-		return -EIO;
+		return PTR_ERR(dir);
 	}
 
 	ret = btrfs_add_link(trans, dir, inode, name, 1, index);
@@ -1844,9 +1837,9 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
 	bool update_size = true;
 	bool name_added = false;
 
-	dir = read_one_inode(root, key->objectid);
-	if (!dir)
-		return -EIO;
+	dir = btrfs_iget_logging(key->objectid, root);
+	if (IS_ERR(dir))
+		return PTR_ERR(dir);
 
 	ret = read_alloc_one_name(eb, di + 1, btrfs_dir_name_len(eb, di), &name);
 	if (ret)
@@ -2146,9 +2139,10 @@ static noinline int check_item_in_log(struct btrfs_trans_handle *trans,
 	btrfs_dir_item_key_to_cpu(eb, di, &location);
 	btrfs_release_path(path);
 	btrfs_release_path(log_path);
-	inode = read_one_inode(root, location.objectid);
-	if (!inode) {
-		ret = -EIO;
+	inode = btrfs_iget_logging(location.objectid, root);
+	if (IS_ERR(inode)) {
+		ret = PTR_ERR(inode);
+		inode = NULL;
 		goto out;
 	}
 
@@ -2300,14 +2294,17 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
 	if (!log_path)
 		return -ENOMEM;
 
-	dir = read_one_inode(root, dirid);
-	/* it isn't an error if the inode isn't there, that can happen
-	 * because we replay the deletes before we copy in the inode item
-	 * from the log
+	dir = btrfs_iget_logging(dirid, root);
+	/*
+	 * It isn't an error if the inode isn't there, that can happen because
+	 * we replay the deletes before we copy in the inode item from the log.
 	 */
-	if (!dir) {
+	if (IS_ERR(dir)) {
 		btrfs_free_path(log_path);
-		return 0;
+		ret = PTR_ERR(dir);
+		if (ret == -ENOENT)
+			ret = 0;
+		return ret;
 	}
 
 	range_start = 0;
@@ -2466,9 +2463,9 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 				struct btrfs_inode *inode;
 				u64 from;
 
-				inode = read_one_inode(root, key.objectid);
-				if (!inode) {
-					ret = -EIO;
+				inode = btrfs_iget_logging(key.objectid, root);
+				if (IS_ERR(inode)) {
+					ret = PTR_ERR(inode);
 					break;
 				}
 				from = ALIGN(i_size_read(&inode->vfs_inode),
Signature · 1 line
2.39.5
author maintainer
#68 lore raw

PATCH 68/178 btrfs: record new subvolume in parent dir earlier to avoid dir logging races

In reply to Greg Kroah-Hartman

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

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

From: Filipe Manana <[email protected]>

[ Upstream commit bf5bcf9a6fa070ec8a725b08db63fb1318f77366 ]

Instead of recording that a new subvolume was created in a directory after we add the entry do the directory, record it before adding the entry. This is to avoid races where after creating the entry and before recording the new subvolume in the directory (the call to btrfs_record_new_subvolume()), another task logs the directory, so we end up with a log tree where we logged a directory that has an entry pointing to a root that was not yet committed, resulting in an invalid entry if the log is persisted and replayed later due to a power failure or crash.

Also state this requirement in the function comment for btrfs_record_new_subvolume(), similar to what we do for the btrfs_record_unlink_dir() and btrfs_record_snapshot_destroy().

Fixes 45c4102f0d82 ("btrfs: avoid transaction commit on any fsync after subvolume creation")
---
 fs/btrfs/ioctl.c    | 4 ++--
 fs/btrfs/tree-log.c | 2 ++
 2 files changed, 4 insertions(+), 2 deletions(-)
fs/btrfs/ioctl.c +2 -2
index 63aeacc549457..b70ef4455610a 100644
@@ -666,14 +666,14 @@ static noinline int create_subvol(struct mnt_idmap *idmap,
 		goto out;
 	}
 
+	btrfs_record_new_subvolume(trans, BTRFS_I(dir));
+
 	ret = btrfs_create_new_inode(trans, &new_inode_args);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
-	btrfs_record_new_subvolume(trans, BTRFS_I(dir));
-
 	d_instantiate_new(dentry, new_inode_args.inode);
 	new_inode_args.inode = NULL;
 
fs/btrfs/tree-log.c +2 -0
index f8611ec11da9d..e05140ce95be9 100644
@@ -7444,6 +7444,8 @@ void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans,
  * full log sync.
  * Also we don't need to worry with renames, since btrfs_rename() marks the log
  * for full commit when renaming a subvolume.
+ *
+ * Must be called before creating the subvolume entry in its parent directory.
  */
 void btrfs_record_new_subvolume(const struct btrfs_trans_handle *trans,
 				struct btrfs_inode *dir)
Signature · 1 line
2.39.5
author maintainer
#41 lore raw

PATCH 41/178 arm64: dts: apple: Move touchbar mipi {address,size}-cells from dtsi to dts

In reply to Greg Kroah-Hartman

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

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

From: Sven Peter <[email protected]>

[ Upstream commit 08a0d93c353bd55de8b5fb77b464d89425be0215 ]

Move the {address,size}-cells property from the (disabled) touchbar screen mipi node inside the dtsi file to the model-specific dts file where it's enabled to fix the following W=1 warnings:

t8103.dtsi:404.34-433.5: Warning (avoid_unnecessary_addr_size): /soc/dsi@228600000: unnecessary #address-cells/#size-cells without "ranges", "dma-ranges" or child "reg" property t8112.dtsi:419.34-448.5: Warning (avoid_unnecessary_addr_size): /soc/dsi@228600000: unnecessary #address-cells/#size-cells without "ranges", "dma-ranges" or child "reg" property

Fixes 7275e795e520 ("arm64: dts: apple: Add touchbar screen nodes")
---
 arch/arm64/boot/dts/apple/t8103-j293.dts | 2 ++
 arch/arm64/boot/dts/apple/t8103.dtsi     | 2 --
 arch/arm64/boot/dts/apple/t8112-j493.dts | 2 ++
 arch/arm64/boot/dts/apple/t8112.dtsi     | 2 --
 4 files changed, 4 insertions(+), 4 deletions(-)
arch/arm64/boot/dts/apple/t8103-j293.dts +2 -0
index e2d9439397f71..5b3c42e9f0e67 100644
@@ -100,6 +100,8 @@ dfr_mipi_out_panel: endpoint@0 {
 
 &displaydfr_mipi {
 	status = "okay";
+	#address-cells = <1>;
+	#size-cells = <0>;
 
 	dfr_panel: panel@0 {
 		compatible = "apple,j293-summit", "apple,summit";
arch/arm64/boot/dts/apple/t8103.dtsi +0 -2
index 97b6a067394e3..229b10efaab9e 100644
@@ -404,8 +404,6 @@ displaydfr_mipi: dsi@228600000 {
 			compatible = "apple,t8103-display-pipe-mipi", "apple,h7-display-pipe-mipi";
 			reg = <0x2 0x28600000 0x0 0x100000>;
 			power-domains = <&ps_mipi_dsi>;
-			#address-cells = <1>;
-			#size-cells = <0>;
 			status = "disabled";
 
 			ports {
arch/arm64/boot/dts/apple/t8112-j493.dts +2 -0
index be86d34c6696c..fb8ad7d4c65a8 100644
@@ -63,6 +63,8 @@ dfr_mipi_out_panel: endpoint@0 {
 
 &displaydfr_mipi {
 	status = "okay";
+	#address-cells = <1>;
+	#size-cells = <0>;
 
 	dfr_panel: panel@0 {
 		compatible = "apple,j493-summit", "apple,summit";
arch/arm64/boot/dts/apple/t8112.dtsi +0 -2
index d9b966d68e4fa..7488e3850493b 100644
@@ -420,8 +420,6 @@ displaydfr_mipi: dsi@228600000 {
 			compatible = "apple,t8112-display-pipe-mipi", "apple,h7-display-pipe-mipi";
 			reg = <0x2 0x28600000 0x0 0x100000>;
 			power-domains = <&ps_mipi_dsi>;
-			#address-cells = <1>;
-			#size-cells = <0>;
 			status = "disabled";
 
 			ports {
Signature · 1 line
2.39.5
author maintainer
#42 lore raw

PATCH 42/178 platform/mellanox: mlxbf-tmfifo: fix vring_desc.len assignment

In reply to Greg Kroah-Hartman

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

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

From: David Thompson <[email protected]>

[ Upstream commit 109f4d29dade8ae5b4ac6325af9d1bc24b4230f8 ]

Fix warnings reported by sparse, related to incorrect type:
drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: warning: incorrect type in assignment (different base types)
drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: expected restricted __virtio32 [usertype] len
drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: got unsigned long

---
 drivers/platform/mellanox/mlxbf-tmfifo.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/platform/mellanox/mlxbf-tmfifo.c +2 -1
index aae99adb29eb0..70c58c4c6c842 100644
@@ -281,7 +281,8 @@ static int mlxbf_tmfifo_alloc_vrings(struct mlxbf_tmfifo *fifo,
 		vring->align = SMP_CACHE_BYTES;
 		vring->index = i;
 		vring->vdev_id = tm_vdev->vdev.id.device;
-		vring->drop_desc.len = VRING_DROP_DESC_MAX_LEN;
+		vring->drop_desc.len = cpu_to_virtio32(&tm_vdev->vdev,
+						       VRING_DROP_DESC_MAX_LEN);
 		dev = &tm_vdev->vdev.dev;
 
 		size = vring_size(vring->num, vring->align);
Signature · 1 line
2.39.5
author maintainer
#43 lore raw

PATCH 43/178 RDMA/mlx5: Fix unsafe xarray access in implicit ODP handling

In reply to Greg Kroah-Hartman

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

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

From: Or Har-Toov <[email protected]>

[ Upstream commit 2c6b640ea08bff1a192bf87fa45246ff1e40767c ]

__xa_store() and __xa_erase() were used without holding the proper lock, which led to a lockdep warning due to unsafe RCU usage. This patch replaces them with xa_store() and xa_erase(), which perform the necessary locking internally.

=============================
WARNING: suspicious RCPU usage 6.14.0-rc7_for_upstream_debug_2025_03_18_15_01 #1 Not tainted
-----------------------------
./include/linux/xarray.h:1211 suspicious rcu_dereference_protected() usage!

other info that might help us debug this:

rcu_scheduler_active = 2, debug_locks = 1
3 locks held by kworker/u136:0/219:

      at: process_one_work+0xbe4/0x15f0
      process_one_work+0x75c/0x15f0
      pagefault_mr+0x9a5/0x1390 [mlx5_ib]
  stack backtrace:
  CPU: 14 UID: 0 PID: 219 Comm: kworker/u136:0 Not tainted
  6.14.0-rc7_for_upstream_debug_2025_03_18_15_01 #1
  Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS
  rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
  Workqueue: mlx5_ib_page_fault mlx5_ib_eqe_pf_action [mlx5_ib]
  Call Trace:
   dump_stack_lvl+0xa8/0xc0
   lockdep_rcu_suspicious+0x1e6/0x260
   xas_create+0xb8a/0xee0
   xas_store+0x73/0x14c0
   __xa_store+0x13c/0x220
   ? xa_store_range+0x390/0x390
   ? spin_bug+0x1d0/0x1d0
   pagefault_mr+0xcb5/0x1390 [mlx5_ib]
   ? _raw_spin_unlock+0x1f/0x30
   mlx5_ib_eqe_pf_action+0x3be/0x2620 [mlx5_ib]
   ? lockdep_hardirqs_on_prepare+0x400/0x400
   ? mlx5_ib_invalidate_range+0xcb0/0xcb0 [mlx5_ib]
   process_one_work+0x7db/0x15f0
   ? pwq_dec_nr_in_flight+0xda0/0xda0
   ? assign_work+0x168/0x240
   worker_thread+0x57d/0xcd0
   ? rescuer_thread+0xc40/0xc40
   kthread+0x3b3/0x800
   ? kthread_is_per_cpu+0xb0/0xb0
   ? lock_downgrade+0x680/0x680
   ? do_raw_spin_lock+0x12d/0x270
   ? spin_bug+0x1d0/0x1d0
   ? finish_task_switch.isra.0+0x284/0x9e0
   ? lockdep_hardirqs_on_prepare+0x284/0x400
   ? kthread_is_per_cpu+0xb0/0xb0
   ret_from_fork+0x2d/0x70
   ? kthread_is_per_cpu+0xb0/0xb0
   ret_from_fork_asm+0x11/0x20
---
 drivers/infiniband/hw/mlx5/odp.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/infiniband/hw/mlx5/odp.c +4 -4
index 86d8fa63bf691..4183bab753a39 100644
@@ -247,8 +247,8 @@ static void destroy_unused_implicit_child_mr(struct mlx5_ib_mr *mr)
 	}
 
 	if (MLX5_CAP_ODP(mr_to_mdev(mr)->mdev, mem_page_fault))
-		__xa_erase(&mr_to_mdev(mr)->odp_mkeys,
-			   mlx5_base_mkey(mr->mmkey.key));
+		xa_erase(&mr_to_mdev(mr)->odp_mkeys,
+			 mlx5_base_mkey(mr->mmkey.key));
 	xa_unlock(&imr->implicit_children);
 
 	/* Freeing a MR is a sleeping operation, so bounce to a work queue */
@@ -521,8 +521,8 @@ static struct mlx5_ib_mr *implicit_get_child_mr(struct mlx5_ib_mr *imr,
 	}
 
 	if (MLX5_CAP_ODP(dev->mdev, mem_page_fault)) {
-		ret = __xa_store(&dev->odp_mkeys, mlx5_base_mkey(mr->mmkey.key),
-				 &mr->mmkey, GFP_KERNEL);
+		ret = xa_store(&dev->odp_mkeys, mlx5_base_mkey(mr->mmkey.key),
+			       &mr->mmkey, GFP_KERNEL);
 		if (xa_is_err(ret)) {
 			ret = ERR_PTR(xa_err(ret));
 			__xa_erase(&imr->implicit_children, idx);
Signature · 1 line
2.39.5
author maintainer
#44 lore raw

PATCH 44/178 RDMA/mlx5: Initialize obj_event->obj_sub_list before xa_insert

In reply to Greg Kroah-Hartman

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

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

From: Mark Zhang <[email protected]>

[ Upstream commit 8edab8a72d67742f87e9dc2e2b0cdfddda5dc29a ]

The obj_event may be loaded immediately after inserted, then if the list_head is not initialized then we may get a poisonous pointer. This fixes the crash below:

mlx5_core 0000:03:00.0: MLX5E: StrdRq(1) RqSz(8) StrdSz(2048) RxCqeCmprss(0 enhanced) mlx5_core.sf mlx5_core.sf.4: firmware version: 32.38.3056
mlx5_core 0000:03:00.0 en3f0pf0sf2002: renamed from eth0
mlx5_core.sf mlx5_core.sf.4: Rate limit: 127 rates are supported, range: 0Mbps to 195312Mbps
IPv6: ADDRCONF(NETDEV_CHANGE): en3f0pf0sf2002: link becomes ready

 Unable to handle kernel NULL pointer dereference at virtual address 0000000000000060
 Mem abort info:
   ESR = 0x96000006
   EC = 0x25: DABT (current EL), IL = 32 bits
   SET = 0, FnV = 0
   EA = 0, S1PTW = 0
 Data abort info:
   ISV = 0, ISS = 0x00000006
   CM = 0, WnR = 0
 user pgtable: 4k pages, 48-bit VAs, pgdp=00000007760fb000
 [0000000000000060] pgd=000000076f6d7003, p4d=000000076f6d7003, pud=0000000777841003, pmd=0000000000000000
 Internal error: Oops: 96000006 [#1] SMP
 Modules linked in: ipmb_host(OE) act_mirred(E) cls_flower(E) sch_ingress(E) mptcp_diag(E) udp_diag(E) raw_diag(E) unix_diag(E) tcp_diag(E) inet_diag(E) binfmt_misc(E) bonding(OE) rdma_ucm(OE) rdma_cm(OE) iw_cm(OE) ib_ipoib(OE) ib_cm(OE) isofs(E) cdrom(E) mst_pciconf(OE) ib_umad(OE) mlx5_ib(OE) ipmb_dev_int(OE) mlx5_core(OE) kpatch_15237886(OEK) mlxdevm(OE) auxiliary(OE) ib_uverbs(OE) ib_core(OE) psample(E) mlxfw(OE) tls(E) sunrpc(E) vfat(E) fat(E) crct10dif_ce(E) ghash_ce(E) sha1_ce(E) sbsa_gwdt(E) virtio_console(E) ext4(E) mbcache(E) jbd2(E) xfs(E) libcrc32c(E) mmc_block(E) virtio_net(E) net_failover(E) failover(E) sha2_ce(E) sha256_arm64(E) nvme(OE) nvme_core(OE) gpio_mlxbf3(OE) mlx_compat(OE) mlxbf_pmc(OE) i2c_mlxbf(OE) sdhci_of_dwcmshc(OE) pinctrl_mlxbf3(OE) mlxbf_pka(OE) gpio_generic(E) i2c_core(E) mmc_core(E) mlxbf_gige(OE) vitesse(E) pwr_mlxbf(OE) mlxbf_tmfifo(OE) micrel(E) mlxbf_bootctl(OE) virtio_ring(E) virtio(E) ipmi_devintf(E) ipmi_msghandler(E)
  [last unloaded: mst_pci]
 CPU: 11 PID: 20913 Comm: rte-worker-11 Kdump: loaded Tainted: G           OE K   5.10.134-13.1.an8.aarch64 #1
 Hardware name: https://www.mellanox.com BlueField-3 SmartNIC Main Card/BlueField-3 SmartNIC Main Card, BIOS 4.2.2.12968 Oct 26 2023
 pstate: a0400089 (NzCv daIf +PAN -UAO -TCO BTYPE=--)
 pc : dispatch_event_fd+0x68/0x300 [mlx5_ib]
 lr : devx_event_notifier+0xcc/0x228 [mlx5_ib]
 sp : ffff80001005bcf0
 x29: ffff80001005bcf0 x28: 0000000000000001
 x27: ffff244e0740a1d8 x26: ffff244e0740a1d0
 x25: ffffda56beff5ae0 x24: ffffda56bf911618
 x23: ffff244e0596a480 x22: ffff244e0596a480
 x21: ffff244d8312ad90 x20: ffff244e0596a480
 x19: fffffffffffffff0 x18: 0000000000000000
 x17: 0000000000000000 x16: ffffda56be66d620
 x15: 0000000000000000 x14: 0000000000000000
 x13: 0000000000000000 x12: 0000000000000000
 x11: 0000000000000040 x10: ffffda56bfcafb50
 x9 : ffffda5655c25f2c x8 : 0000000000000010
 x7 : 0000000000000000 x6 : ffff24545a2e24b8
 x5 : 0000000000000003 x4 : ffff80001005bd28
 x3 : 0000000000000000 x2 : 0000000000000000
 x1 : ffff244e0596a480 x0 : ffff244d8312ad90
 Call trace:
  dispatch_event_fd+0x68/0x300 [mlx5_ib]
  devx_event_notifier+0xcc/0x228 [mlx5_ib]
  atomic_notifier_call_chain+0x58/0x80
  mlx5_eq_async_int+0x148/0x2b0 [mlx5_core]
  atomic_notifier_call_chain+0x58/0x80
  irq_int_handler+0x20/0x30 [mlx5_core]
  __handle_irq_event_percpu+0x60/0x220
  handle_irq_event_percpu+0x3c/0x90
  handle_irq_event+0x58/0x158
  handle_fasteoi_irq+0xfc/0x188
  generic_handle_irq+0x34/0x48
  ...
---
 drivers/infiniband/hw/mlx5/devx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/hw/mlx5/devx.c +1 -1
index bceae1c1f9801..843dcd3122424 100644
@@ -1958,6 +1958,7 @@ subscribe_event_xa_alloc(struct mlx5_devx_event_table *devx_event_table,
 			/* Level1 is valid for future use, no need to free */
 			return -ENOMEM;
 
+		INIT_LIST_HEAD(&obj_event->obj_sub_list);
 		err = xa_insert(&event->object_ids,
 				key_level2,
 				obj_event,
@@ -1966,7 +1967,6 @@ subscribe_event_xa_alloc(struct mlx5_devx_event_table *devx_event_table,
 			kfree(obj_event);
 			return err;
 		}
-		INIT_LIST_HEAD(&obj_event->obj_sub_list);
 	}
 
 	return 0;
Signature · 1 line
2.39.5
author maintainer
#45 lore raw

PATCH 45/178 module: Provide EXPORT_SYMBOL_GPL_FOR_MODULES() helper

In reply to Greg Kroah-Hartman

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

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

From: Peter Zijlstra <[email protected]>

[ Upstream commit 707f853d7fa3ce323a6875487890c213e34d81a0 ]

Helper macro to more easily limit the export of a symbol to a given list of modules.

Eg:

  EXPORT_SYMBOL_GPL_FOR_MODULES(preempt_notifier_inc, "kvm");

will limit the use of said function to kvm.ko, any other module trying to use this symbol will refure to load (and get modpost build failures).

Stable-dep-of: cbe4134ea4bc ("fs: export anon_inode_make_secure_inode() and fix secretmem LSM bypass")

---
 Documentation/core-api/symbol-namespaces.rst | 22 ++++++++++++++++++++
 include/linux/export.h                       | 12 +++++++++--
 2 files changed, 32 insertions(+), 2 deletions(-)
Documentation/core-api/symbol-namespaces.rst +22 -0
index 06f766a6aab24..c6f59c5e25648 100644
@@ -28,6 +28,9 @@ kernel. As of today, modules that make use of symbols exported into namespaces,
 are required to import the namespace. Otherwise the kernel will, depending on
 its configuration, reject loading the module or warn about a missing import.
 
+Additionally, it is possible to put symbols into a module namespace, strictly
+limiting which modules are allowed to use these symbols.
+
 2. How to define Symbol Namespaces
 ==================================
 
@@ -83,6 +86,22 @@ unit as preprocessor statement. The above example would then read::
 within the corresponding compilation unit before the #include for
 <linux/export.h>. Typically it's placed before the first #include statement.
 
+2.3 Using the EXPORT_SYMBOL_GPL_FOR_MODULES() macro
+===================================================
+
+Symbols exported using this macro are put into a module namespace. This
+namespace cannot be imported.
+
+The macro takes a comma separated list of module names, allowing only those
+modules to access this symbol. Simple tail-globs are supported.
+
+For example:
+
+  EXPORT_SYMBOL_GPL_FOR_MODULES(preempt_notifier_inc, "kvm,kvm-*")
+
+will limit usage of this symbol to modules whoes name matches the given
+patterns.
+
 3. How to use Symbols exported in Namespaces
 ============================================
 
@@ -154,3 +173,6 @@ in-tree modules::
 You can also run nsdeps for external module builds. A typical usage is::
 
 	$ make -C <path_to_kernel_src> M=$PWD nsdeps
+
+Note: it will happily generate an import statement for the module namespace;
+which will not work and generates build and runtime failures.
include/linux/export.h +10 -2
index a8c23d945634b..f35d03b4113b1 100644
@@ -24,11 +24,17 @@
 	.long sym
 #endif
 
-#define ___EXPORT_SYMBOL(sym, license, ns)		\
+/*
+ * LLVM integrated assembler cam merge adjacent string literals (like
+ * C and GNU-as) passed to '.ascii', but not to '.asciz' and chokes on:
+ *
+ *   .asciz "MODULE_" "kvm" ;
+ */
+#define ___EXPORT_SYMBOL(sym, license, ns...)		\
 	.section ".export_symbol","a"		ASM_NL	\
 	__export_symbol_##sym:			ASM_NL	\
 		.asciz license			ASM_NL	\
-		.asciz ns			ASM_NL	\
+		.ascii ns "\0"			ASM_NL	\
 		__EXPORT_SYMBOL_REF(sym)	ASM_NL	\
 	.previous
 
@@ -85,4 +91,6 @@
 #define EXPORT_SYMBOL_NS(sym, ns)	__EXPORT_SYMBOL(sym, "", ns)
 #define EXPORT_SYMBOL_NS_GPL(sym, ns)	__EXPORT_SYMBOL(sym, "GPL", ns)
 
+#define EXPORT_SYMBOL_GPL_FOR_MODULES(sym, mods) __EXPORT_SYMBOL(sym, "GPL", "module:" mods)
+
 #endif /* _LINUX_EXPORT_H */
Signature · 1 line
2.39.5
author maintainer
#46 lore raw

PATCH 46/178 fs: export anon_inode_make_secure_inode() and fix secretmem LSM bypass

In reply to Greg Kroah-Hartman

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

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

From: Shivank Garg <[email protected]>

[ Upstream commit cbe4134ea4bc493239786220bd69cb8a13493190 ]

Export anon_inode_make_secure_inode() to allow KVM guest_memfd to create anonymous inodes with proper security context. This replaces the current pattern of calling alloc_anon_inode() followed by inode_init_security_anon() for creating security context manually.

This change also fixes a security regression in secretmem where the S_PRIVATE flag was not cleared after alloc_anon_inode(), causing LSM/SELinux checks to be bypassed for secretmem file descriptors.

As guest_memfd currently resides in the KVM module, we need to export this symbol for use outside the core kernel. In the future, guest_memfd might be moved to core-mm, at which point the symbols no longer would have to be exported. When/if that happens is still unclear.

---
 fs/anon_inodes.c   | 23 ++++++++++++++++++-----
 include/linux/fs.h |  2 ++
 mm/secretmem.c     |  9 +--------
 3 files changed, 21 insertions(+), 13 deletions(-)
fs/anon_inodes.c +18 -5
index e51e7d88980a2..1d847a939f29a 100644
@@ -98,14 +98,25 @@ static struct file_system_type anon_inode_fs_type = {
 	.kill_sb	= kill_anon_super,
 };
 
-static struct inode *anon_inode_make_secure_inode(
-	const char *name,
-	const struct inode *context_inode)
+/**
+ * anon_inode_make_secure_inode - allocate an anonymous inode with security context
+ * @sb:		[in]	Superblock to allocate from
+ * @name:	[in]	Name of the class of the newfile (e.g., "secretmem")
+ * @context_inode:
+ *		[in]	Optional parent inode for security inheritance
+ *
+ * The function ensures proper security initialization through the LSM hook
+ * security_inode_init_security_anon().
+ *
+ * Return:	Pointer to new inode on success, ERR_PTR on failure.
+ */
+struct inode *anon_inode_make_secure_inode(struct super_block *sb, const char *name,
+					   const struct inode *context_inode)
 {
 	struct inode *inode;
 	int error;
 
-	inode = alloc_anon_inode(anon_inode_mnt->mnt_sb);
+	inode = alloc_anon_inode(sb);
 	if (IS_ERR(inode))
 		return inode;
 	inode->i_flags &= ~S_PRIVATE;
@@ -118,6 +129,7 @@ static struct inode *anon_inode_make_secure_inode(
 	}
 	return inode;
 }
+EXPORT_SYMBOL_GPL_FOR_MODULES(anon_inode_make_secure_inode, "kvm");
 
 static struct file *__anon_inode_getfile(const char *name,
 					 const struct file_operations *fops,
@@ -132,7 +144,8 @@ static struct file *__anon_inode_getfile(const char *name,
 		return ERR_PTR(-ENOENT);
 
 	if (make_inode) {
-		inode =	anon_inode_make_secure_inode(name, context_inode);
+		inode =	anon_inode_make_secure_inode(anon_inode_mnt->mnt_sb,
+						     name, context_inode);
 		if (IS_ERR(inode)) {
 			file = ERR_CAST(inode);
 			goto err;
include/linux/fs.h +2 -0
index ef41a8213a6fa..1fe7bf10d442c 100644
@@ -3552,6 +3552,8 @@ extern int simple_write_begin(struct file *file, struct address_space *mapping,
 extern const struct address_space_operations ram_aops;
 extern int always_delete_dentry(const struct dentry *);
 extern struct inode *alloc_anon_inode(struct super_block *);
+struct inode *anon_inode_make_secure_inode(struct super_block *sb, const char *name,
+					   const struct inode *context_inode);
 extern int simple_nosetlease(struct file *, int, struct file_lease **, void **);
 extern const struct dentry_operations simple_dentry_operations;
 
mm/secretmem.c +1 -8
index 1b0a214ee5580..4662f2510ae5f 100644
@@ -195,18 +195,11 @@ static struct file *secretmem_file_create(unsigned long flags)
 	struct file *file;
 	struct inode *inode;
 	const char *anon_name = "[secretmem]";
-	int err;
 
-	inode = alloc_anon_inode(secretmem_mnt->mnt_sb);
+	inode = anon_inode_make_secure_inode(secretmem_mnt->mnt_sb, anon_name, NULL);
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
 
-	err = security_inode_init_security_anon(inode, &QSTR(anon_name), NULL);
-	if (err) {
-		file = ERR_PTR(err);
-		goto err_free_inode;
-	}
-
 	file = alloc_file_pseudo(inode, secretmem_mnt, "secretmem",
 				 O_RDWR, &secretmem_fops);
 	if (IS_ERR(file))
Signature · 1 line
2.39.5
author maintainer
#37 lore raw

PATCH 37/178 firmware: exynos-acpm: fix timeouts on xfers handling

In reply to Greg Kroah-Hartman

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

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

From: Tudor Ambarus <[email protected]>

[ Upstream commit 8d2c2fa2209e83d0eb10f7330d8a0bbdc1df32ff ]

The mailbox framework has a single inflight request at a time. If a request is sent while another is still active, it will be queued to the mailbox core ring buffer.

ACPM protocol did not serialize the calls to the mailbox subsystem so we could start the timeout ticks in parallel for multiple requests, while just one was being inflight.

Consider a hypothetical case where the xfer timeout is 100ms and an ACPM transaction takes 90ms:

      | 0ms: Message #0 is queued in mailbox layer and sent out, then sits
      |      at acpm_dequeue_by_polling() with a timeout of 100ms
      | 1ms: Message #1 is queued in mailbox layer but not sent out yet.
      |      Since send_message() doesn't block, it also sits at
      |      acpm_dequeue_by_polling() with a timeout of 100ms
      |  ...
      | 90ms: Message #0 is completed, txdone is called and message #1 is sent
      | 101ms: Message #1 times out since the count started at 1ms. Even though
      |       it has only been inflight for 11ms.

Fix the problem by moving mbox_send_message() and mbox_client_txdone() immediately after the message has been written to the TX queue and while still keeping the ACPM TX queue lock. We thus tie together the TX write with the doorbell ring and mark the TX as done after the doorbell has been rung. This guarantees that the doorbell has been rang before starting the timeout ticks. We should also see some performance improvement as we no longer wait to receive a response before ringing the doorbell for the next request, so the ACPM firmware shall be able to drain faster the TX queue. Another benefit is that requests are no longer able to ring the doorbell one for the other, so it eases debugging. Finally, the mailbox software queue will always contain a single doorbell request due to the serialization done at the ACPM TX queue level. Protocols like ACPM, that handle their own hardware queues need a passthrough mailbox API, where they are able to just ring the doorbell or flip a bit directly into the mailbox controller. The mailbox software queue mechanism, the locking done into the mailbox core is not really needed, so hopefully this lays the foundation for a passthrough mailbox API.

---
 drivers/firmware/samsung/exynos-acpm.c | 25 +++++++++----------------
 1 file changed, 9 insertions(+), 16 deletions(-)
drivers/firmware/samsung/exynos-acpm.c +9 -16
index e80cb7a8da8f2..520a9fd3b0fd3 100644
@@ -430,6 +430,9 @@ int acpm_do_xfer(const struct acpm_handle *handle, const struct acpm_xfer *xfer)
 		return -EOPNOTSUPP;
 	}
 
+	msg.chan_id = xfer->acpm_chan_id;
+	msg.chan_type = EXYNOS_MBOX_CHAN_TYPE_DOORBELL;
+
 	scoped_guard(mutex, &achan->tx_lock) {
 		tx_front = readl(achan->tx.front);
 		idx = (tx_front + 1) % achan->qlen;
@@ -446,25 +449,15 @@ int acpm_do_xfer(const struct acpm_handle *handle, const struct acpm_xfer *xfer)
 
 		/* Advance TX front. */
 		writel(idx, achan->tx.front);
-	}
 
-	msg.chan_id = xfer->acpm_chan_id;
-	msg.chan_type = EXYNOS_MBOX_CHAN_TYPE_DOORBELL;
-	ret = mbox_send_message(achan->chan, (void *)&msg);
-	if (ret < 0)
-		return ret;
-
-	ret = acpm_wait_for_message_response(achan, xfer);
+		ret = mbox_send_message(achan->chan, (void *)&msg);
+		if (ret < 0)
+			return ret;
 
-	/*
-	 * NOTE: we might prefer not to need the mailbox ticker to manage the
-	 * transfer queueing since the protocol layer queues things by itself.
-	 * Unfortunately, we have to kick the mailbox framework after we have
-	 * received our message.
-	 */
-	mbox_client_txdone(achan->chan, ret);
+		mbox_client_txdone(achan->chan, 0);
+	}
 
-	return ret;
+	return acpm_wait_for_message_response(achan, xfer);
 }
 
 /**
Signature · 1 line
2.39.5
author maintainer
#70 lore raw

PATCH 70/178 btrfs: use btrfs_record_snapshot_destroy() during rmdir

In reply to Greg Kroah-Hartman

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

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

From: Filipe Manana <[email protected]>

[ Upstream commit 157501b0469969fc1ba53add5049575aadd79d80 ]

We are setting the parent directory's last_unlink_trans directly which may result in a concurrent task starting to log the directory not see the update and therefore can log the directory after we removed a child directory which had a snapshot within instead of falling back to a transaction commit. Replaying such a log tree would result in a mount failure since we can't currently delete snapshots (and subvolumes) during log replay. This is the type of failure described in commit 1ec9a1ae1e30 ("Btrfs: fix unreplayable log after snapshot delete + parent dir fsync").

Fix this by using btrfs_record_snapshot_destroy() which updates the last_unlink_trans field while holding the inode's log_mutex lock.

Fixes 44f714dae50a ("Btrfs: improve performance on fsync against new inode after rename/unlink")
---
 fs/btrfs/inode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/inode.c +1 -1
index 24db00a9319b3..f18f4d59d389e 100644
@@ -4764,7 +4764,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 	 * deletes for directory foo.
 	 */
 	if (BTRFS_I(inode)->last_unlink_trans >= trans->transid)
-		BTRFS_I(dir)->last_unlink_trans = BTRFS_I(inode)->last_unlink_trans;
+		btrfs_record_snapshot_destroy(trans, BTRFS_I(dir));
 
 	if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
 		ret = btrfs_unlink_subvol(trans, BTRFS_I(dir), dentry);
Signature · 1 line
2.39.5
author maintainer
#79 lore raw

PATCH 79/178 nvme-pci: refresh visible attrs after being checked

In reply to Greg Kroah-Hartman

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

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

From: Eugen Hristev <[email protected]>

[ Upstream commit 14005c96d6649b27fa52d0cd0f492eb3b5586c07 ]

The sysfs attributes are registered early, but the driver does not know whether they are needed or not at that moment.

For the CMB attributes, commit e917a849c3fc ("nvme-pci: refresh visible attrs for cmb attributes") solved this problem by
calling nvme_update_attrs after mapping the CMB. However the issue persists for the HMB attributes. To solve the problem, moved the call to nvme_update_attrs after nvme_setup_host_mem, which sets up the HMB.

Fixes e917a849c3fc ("nvme-pci: refresh visible attrs for cmb attributes")
Fixes 86adbf0cdb9e ("nvme: simplify transport specific device attribute handling")
---
 drivers/nvme/host/pci.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/nvme/host/pci.c +4 -2
index f1dd804151b1c..776c867fb64d5 100644
@@ -2031,8 +2031,6 @@ static void nvme_map_cmb(struct nvme_dev *dev)
 	if ((dev->cmbsz & (NVME_CMBSZ_WDS | NVME_CMBSZ_RDS)) ==
 			(NVME_CMBSZ_WDS | NVME_CMBSZ_RDS))
 		pci_p2pmem_publish(pdev, true);
-
-	nvme_update_attrs(dev);
 }
 
 static int nvme_set_host_mem(struct nvme_dev *dev, u32 bits)
@@ -2969,6 +2967,8 @@ static void nvme_reset_work(struct work_struct *work)
 	if (result < 0)
 		goto out;
 
+	nvme_update_attrs(dev);
+
 	result = nvme_setup_io_queues(dev);
 	if (result)
 		goto out;
@@ -3305,6 +3305,8 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (result < 0)
 		goto out_disable;
 
+	nvme_update_attrs(dev);
+
 	result = nvme_setup_io_queues(dev);
 	if (result)
 		goto out_disable;
Signature · 1 line
2.39.5
author maintainer
#80 lore raw

PATCH 80/178 platform/x86: dell-wmi-sysman: Fix WMI data block retrieval in sysfs callbacks

In reply to Greg Kroah-Hartman

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

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

From: Kurt Borja <[email protected]>

[ Upstream commit eb617dd25ca176f3fee24f873f0fd60010773d67 ]

After retrieving WMI data blocks in sysfs callbacks, check for the validity of them before dereferencing their content.

---
 .../platform/x86/dell/dell-wmi-sysman/dell-wmi-sysman.h   | 5 +++++
 .../platform/x86/dell/dell-wmi-sysman/enum-attributes.c   | 5 +++--
 .../platform/x86/dell/dell-wmi-sysman/int-attributes.c    | 5 +++--
 .../x86/dell/dell-wmi-sysman/passobj-attributes.c         | 5 +++--
 .../platform/x86/dell/dell-wmi-sysman/string-attributes.c | 5 +++--
 drivers/platform/x86/dell/dell-wmi-sysman/sysman.c        | 8 ++++----
 6 files changed, 21 insertions(+), 12 deletions(-)
drivers/platform/x86/dell/dell-wmi-sysman/dell-wmi-sysman.h +5 -0
index 3ad33a094588c..817ee7ba07ca0 100644
@@ -89,6 +89,11 @@ extern struct wmi_sysman_priv wmi_priv;
 
 enum { ENUM, INT, STR, PO };
 
+#define ENUM_MIN_ELEMENTS		8
+#define INT_MIN_ELEMENTS		9
+#define STR_MIN_ELEMENTS		8
+#define PO_MIN_ELEMENTS			4
+
 enum {
 	ATTR_NAME,
 	DISPL_NAME_LANG_CODE,
drivers/platform/x86/dell/dell-wmi-sysman/enum-attributes.c +3 -2
index 8cc212c852668..fc2f58b4cbc6e 100644
@@ -23,9 +23,10 @@ static ssize_t current_value_show(struct kobject *kobj, struct kobj_attribute *a
 	obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_ENUMERATION_ATTRIBUTE_GUID);
 	if (!obj)
 		return -EIO;
-	if (obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) {
+	if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < ENUM_MIN_ELEMENTS ||
+	    obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) {
 		kfree(obj);
-		return -EINVAL;
+		return -EIO;
 	}
 	ret = snprintf(buf, PAGE_SIZE, "%s\n", obj->package.elements[CURRENT_VAL].string.pointer);
 	kfree(obj);
drivers/platform/x86/dell/dell-wmi-sysman/int-attributes.c +3 -2
index 951e75b538fad..7352480642391 100644
@@ -25,9 +25,10 @@ static ssize_t current_value_show(struct kobject *kobj, struct kobj_attribute *a
 	obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_INTEGER_ATTRIBUTE_GUID);
 	if (!obj)
 		return -EIO;
-	if (obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_INTEGER) {
+	if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < INT_MIN_ELEMENTS ||
+	    obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_INTEGER) {
 		kfree(obj);
-		return -EINVAL;
+		return -EIO;
 	}
 	ret = snprintf(buf, PAGE_SIZE, "%lld\n", obj->package.elements[CURRENT_VAL].integer.value);
 	kfree(obj);
drivers/platform/x86/dell/dell-wmi-sysman/passobj-attributes.c +3 -2
index d8f1bf5e58a0f..3167e06d416ed 100644
@@ -26,9 +26,10 @@ static ssize_t is_enabled_show(struct kobject *kobj, struct kobj_attribute *attr
 	obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_PASSOBJ_ATTRIBUTE_GUID);
 	if (!obj)
 		return -EIO;
-	if (obj->package.elements[IS_PASS_SET].type != ACPI_TYPE_INTEGER) {
+	if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < PO_MIN_ELEMENTS ||
+	    obj->package.elements[IS_PASS_SET].type != ACPI_TYPE_INTEGER) {
 		kfree(obj);
-		return -EINVAL;
+		return -EIO;
 	}
 	ret = snprintf(buf, PAGE_SIZE, "%lld\n", obj->package.elements[IS_PASS_SET].integer.value);
 	kfree(obj);
drivers/platform/x86/dell/dell-wmi-sysman/string-attributes.c +3 -2
index c392f0ecf8b55..0d2c74f8d1aad 100644
@@ -25,9 +25,10 @@ static ssize_t current_value_show(struct kobject *kobj, struct kobj_attribute *a
 	obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_STRING_ATTRIBUTE_GUID);
 	if (!obj)
 		return -EIO;
-	if (obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) {
+	if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < STR_MIN_ELEMENTS ||
+	    obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) {
 		kfree(obj);
-		return -EINVAL;
+		return -EIO;
 	}
 	ret = snprintf(buf, PAGE_SIZE, "%s\n", obj->package.elements[CURRENT_VAL].string.pointer);
 	kfree(obj);
drivers/platform/x86/dell/dell-wmi-sysman/sysman.c +4 -4
index d00389b860e4e..3c74d5e8350a4 100644
@@ -407,10 +407,10 @@ static int init_bios_attributes(int attr_type, const char *guid)
 		return retval;
 
 	switch (attr_type) {
-	case ENUM:	min_elements = 8;	break;
-	case INT:	min_elements = 9;	break;
-	case STR:	min_elements = 8;	break;
-	case PO:	min_elements = 4;	break;
+	case ENUM:	min_elements = ENUM_MIN_ELEMENTS;	break;
+	case INT:	min_elements = INT_MIN_ELEMENTS;	break;
+	case STR:	min_elements = STR_MIN_ELEMENTS;	break;
+	case PO:	min_elements = PO_MIN_ELEMENTS;		break;
 	default:
 		pr_err("Error: Unknown attr_type: %d\n", attr_type);
 		return -EINVAL;
Signature · 1 line
2.39.5
author maintainer
#81 lore raw

PATCH 81/178 platform/x86: hp-bioscfg: Fix class device unregistration

In reply to Greg Kroah-Hartman

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

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

From: Kurt Borja <[email protected]>

[ Upstream commit 11cba4793b95df3bc192149a6eb044f69aa0b99e ]

Devices under the firmware_attributes_class do not have unique a dev_t. Therefore, device_unregister() should be used instead of device_destroy(), since the latter may match any device with a given dev_t.

---
 drivers/platform/x86/hp/hp-bioscfg/bioscfg.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/platform/x86/hp/hp-bioscfg/bioscfg.c +2 -2
index 13237890fc920..5bfa7159f5bcd 100644
@@ -1034,7 +1034,7 @@ static int __init hp_init(void)
 	release_attributes_data();
 
 err_destroy_classdev:
-	device_destroy(&firmware_attributes_class, MKDEV(0, 0));
+	device_unregister(bioscfg_drv.class_dev);
 
 err_unregister_class:
 	hp_exit_attr_set_interface();
@@ -1045,7 +1045,7 @@ static int __init hp_init(void)
 static void __exit hp_exit(void)
 {
 	release_attributes_data();
-	device_destroy(&firmware_attributes_class, MKDEV(0, 0));
+	device_unregister(bioscfg_drv.class_dev);
 
 	hp_exit_attr_set_interface();
 }
Signature · 1 line
2.39.5
author maintainer
#82 lore raw

PATCH 82/178 platform/x86: think-lmi: Fix class device unregistration

In reply to Greg Kroah-Hartman

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

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

From: Kurt Borja <[email protected]>

[ Upstream commit 5ff1fbb3059730700b4823f43999fc1315984632 ]

Devices under the firmware_attributes_class do not have unique a dev_t. Therefore, device_unregister() should be used instead of device_destroy(), since the latter may match any device with a given dev_t.

Fixes a40cd7ef22fb ("platform/x86: think-lmi: Add WMI interface support on Lenovo platforms")
---
 drivers/platform/x86/think-lmi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/platform/x86/think-lmi.c +2 -2
index 00b1e7c79a3d1..28ce475f6c377 100644
@@ -1554,7 +1554,7 @@ static int tlmi_sysfs_init(void)
 fail_create_attr:
 	tlmi_release_attr();
 fail_device_created:
-	device_destroy(&firmware_attributes_class, MKDEV(0, 0));
+	device_unregister(tlmi_priv.class_dev);
 fail_class_created:
 	return ret;
 }
@@ -1781,7 +1781,7 @@ static int tlmi_analyze(struct wmi_device *wdev)
 static void tlmi_remove(struct wmi_device *wdev)
 {
 	tlmi_release_attr();
-	device_destroy(&firmware_attributes_class, MKDEV(0, 0));
+	device_unregister(tlmi_priv.class_dev);
 }
 
 static int tlmi_probe(struct wmi_device *wdev, const void *context)
Signature · 1 line
2.39.5
author maintainer
#83 lore raw

PATCH 83/178 platform/x86: dell-wmi-sysman: Fix class device unregistration

In reply to Greg Kroah-Hartman

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

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

From: Kurt Borja <[email protected]>

[ Upstream commit 314e5ad4782d08858b3abc325c0487bd2abc23a1 ]

Devices under the firmware_attributes_class do not have unique a dev_t. Therefore, device_unregister() should be used instead of device_destroy(), since the latter may match any device with a given dev_t.

Fixes e8a60aa7404b ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
---
 drivers/platform/x86/dell/dell-wmi-sysman/sysman.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/platform/x86/dell/dell-wmi-sysman/sysman.c +2 -2
index 3c74d5e8350a4..f5402b7146572 100644
@@ -597,7 +597,7 @@ static int __init sysman_init(void)
 	release_attributes_data();
 
 err_destroy_classdev:
-	device_destroy(&firmware_attributes_class, MKDEV(0, 0));
+	device_unregister(wmi_priv.class_dev);
 
 err_exit_bios_attr_pass_interface:
 	exit_bios_attr_pass_interface();
@@ -611,7 +611,7 @@ static int __init sysman_init(void)
 static void __exit sysman_exit(void)
 {
 	release_attributes_data();
-	device_destroy(&firmware_attributes_class, MKDEV(0, 0));
+	device_unregister(wmi_priv.class_dev);
 	exit_bios_attr_set_interface();
 	exit_bios_attr_pass_interface();
 }
Signature · 1 line
2.39.5
author maintainer
#84 lore raw

PATCH 84/178 platform/mellanox: mlxreg-lc: Fix logic error in power state check

In reply to Greg Kroah-Hartman

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

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

From: Alok Tiwari <[email protected]>

[ Upstream commit 644bec18e705ca41d444053407419a21832fcb2f ]

Fixes a logic issue in mlxreg_lc_completion_notify() where the intention was to check if MLXREG_LC_POWERED flag is not set before powering on the device.

The original code used "state & ~MLXREG_LC_POWERED" to check for the absence of the POWERED bit. However this condition evaluates to true even when other bits are set, leading to potentially incorrect behavior.

Corrected the logic to explicitly check for the absence of MLXREG_LC_POWERED using !(state & MLXREG_LC_POWERED).

Fixes 62f9529b8d5c ("platform/mellanox: mlxreg-lc: Add initial support for Nvidia line card devices")
---
 drivers/platform/mellanox/mlxreg-lc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/mellanox/mlxreg-lc.c +1 -1
index aee395bb48ae4..8681ceb7144ba 100644
@@ -688,7 +688,7 @@ static int mlxreg_lc_completion_notify(void *handle, struct i2c_adapter *parent,
 	if (regval & mlxreg_lc->data->mask) {
 		mlxreg_lc->state |= MLXREG_LC_SYNCED;
 		mlxreg_lc_state_update_locked(mlxreg_lc, MLXREG_LC_SYNCED, 1);
-		if (mlxreg_lc->state & ~MLXREG_LC_POWERED) {
+		if (!(mlxreg_lc->state & MLXREG_LC_POWERED)) {
 			err = mlxreg_lc_power_on_off(mlxreg_lc, 1);
 			if (err)
 				goto mlxreg_lc_regmap_power_on_off_fail;
Signature · 1 line
2.39.5
author maintainer
#85 lore raw

PATCH 85/178 drm/bridge: panel: move prepare_prev_first handling to drm_panel_bridge_add_typed

In reply to Greg Kroah-Hartman

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

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

From: Dmitry Baryshkov <[email protected]>

[ Upstream commit eb028cd884e1b0976ff8c5944ee6650fe3ed0a6c ]

The commit 5ea6b1702781 ("drm/panel: Add prepare_prev_first flag to drm_panel") and commit 0974687a19c3 ("drm/bridge: panel: Set pre_enable_prev_first from drmm_panel_bridge_add") added handling of panel's prepare_prev_first to devm_panel_bridge_add() and drmm_panel_bridge_add(). However if the driver calls drm_panel_bridge_add_typed() directly, then the flag won't be handled and thus the drm_bridge.pre_enable_prev_first will not be set.

Move prepare_prev_first handling to the drm_panel_bridge_add_typed() so that there is no way to miss the flag.

Fixes 5ea6b1702781 ("drm/panel: Add prepare_prev_first flag to drm_panel")
Fixes 0974687a19c3 ("drm/bridge: panel: Set pre_enable_prev_first from drmm_panel_bridge_add")
---
 drivers/gpu/drm/bridge/panel.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/gpu/drm/bridge/panel.c +1 -4
index 258c85c83a283..6c58d0759f227 100644
@@ -298,6 +298,7 @@ struct drm_bridge *drm_panel_bridge_add_typed(struct drm_panel *panel,
 	panel_bridge->bridge.of_node = panel->dev->of_node;
 	panel_bridge->bridge.ops = DRM_BRIDGE_OP_MODES;
 	panel_bridge->bridge.type = connector_type;
+	panel_bridge->bridge.pre_enable_prev_first = panel->prepare_prev_first;
 
 	drm_bridge_add(&panel_bridge->bridge);
 
@@ -412,8 +413,6 @@ struct drm_bridge *devm_drm_panel_bridge_add_typed(struct device *dev,
 		return bridge;
 	}
 
-	bridge->pre_enable_prev_first = panel->prepare_prev_first;
-
 	*ptr = bridge;
 	devres_add(dev, ptr);
 
@@ -455,8 +454,6 @@ struct drm_bridge *drmm_panel_bridge_add(struct drm_device *drm,
 	if (ret)
 		return ERR_PTR(ret);
 
-	bridge->pre_enable_prev_first = panel->prepare_prev_first;
-
 	return bridge;
 }
 EXPORT_SYMBOL(drmm_panel_bridge_add);
Signature · 1 line
2.39.5
author maintainer
#86 lore raw

PATCH 86/178 drm/bridge: aux-hpd-bridge: fix assignment of the of_node

In reply to Greg Kroah-Hartman

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

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

From: Dmitry Baryshkov <[email protected]>

[ Upstream commit e8537cad824065b0425fb0429e762e14a08067c2 ]

Perform fix similar to the one in the commit 85e444a68126 ("drm/bridge:
Fix assignment of the of_node of the parent to aux bridge").

The assignment of the of_node to the aux HPD bridge needs to mark the of_node as reused, otherwise driver core will attempt to bind resources like pinctrl, which is going to fail as corresponding pins are already marked as used by the parent device.
Fix that by using the device_set_of_node_from_dev() helper instead of assigning it directly.

---
 drivers/gpu/drm/bridge/aux-hpd-bridge.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/bridge/aux-hpd-bridge.c +2 -1
index 48f297c78ee67..1ec4f16b9939d 100644
@@ -64,10 +64,11 @@ struct auxiliary_device *devm_drm_dp_hpd_bridge_alloc(struct device *parent, str
 	adev->id = ret;
 	adev->name = "dp_hpd_bridge";
 	adev->dev.parent = parent;
-	adev->dev.of_node = of_node_get(parent->of_node);
 	adev->dev.release = drm_aux_hpd_bridge_release;
 	adev->dev.platform_data = of_node_get(np);
 
+	device_set_of_node_from_dev(&adev->dev, parent);
+
 	ret = auxiliary_device_init(adev);
 	if (ret) {
 		of_node_put(adev->dev.platform_data);
Signature · 1 line
2.39.5
author maintainer
#87 lore raw

PATCH 87/178 smb: client: fix warning when reconnecting channel

In reply to Greg Kroah-Hartman

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

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

From: Paulo Alcantara <[email protected]>

[ Upstream commit 3bbe46716092d8ef6b0df4b956f585c5cd0fc78e ]

When reconnecting a channel in smb2_reconnect_server(), a dummy tcon is passed down to smb2_reconnect() with ->query_interface uninitialized, so we can't call queue_delayed_work() on it.

Fix the following warning by ensuring that we're queueing the delayed worker from correct tcon.

WARNING: CPU: 4 PID: 1126 at kernel/workqueue.c:2498 __queue_delayed_work+0x1d2/0x200
Modules linked in: cifs cifs_arc4 nls_ucs2_utils cifs_md4 [last unloaded: cifs]
CPU: 4 UID: 0 PID: 1126 Comm: kworker/4:0 Not tainted 6.16.0-rc3 #5 PREEMPT(voluntary)
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-4.fc42 04/01/2014
Workqueue: cifsiod smb2_reconnect_server [cifs]
RIP: 0010:__queue_delayed_work+0x1d2/0x200
Code: 41 5e 41 5f e9 7f ee ff ff 90 0f 0b 90 e9 5d ff ff ff bf 02 00
00 00 e8 6c f3 07 00 89 c3 eb bd 90 0f 0b 90 e9 57 f> 0b 90 e9 65 fe
ff ff 90 0f 0b 90 e9 72 fe ff ff 90 0f 0b 90 e9
RSP: 0018:ffffc900014afad8 EFLAGS: 00010003
RAX: 0000000000000000 RBX: ffff888124d99988 RCX: ffffffff81399cc1
RDX: dffffc0000000000 RSI: ffff888114326e00 RDI: ffff888124d999f0
RBP: 000000000000ea60 R08: 0000000000000001 R09: ffffed10249b3331
R10: ffff888124d9998f R11: 0000000000000004 R12: 0000000000000040
R13: ffff888114326e00 R14: ffff888124d999d8 R15: ffff888114939020
FS:  0000000000000000(0000) GS:ffff88829f7fe000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007ffe7a2b4038 CR3: 0000000120a6f000 CR4: 0000000000750ef0
PKRU: 55555554
Call Trace:
 <TASK>
 queue_delayed_work_on+0xb4/0xc0
 smb2_reconnect+0xb22/0xf50 [cifs]
 smb2_reconnect_server+0x413/0xd40 [cifs]
 ? __pfx_smb2_reconnect_server+0x10/0x10 [cifs]
 ? local_clock_noinstr+0xd/0xd0
 ? local_clock+0x15/0x30
 ? lock_release+0x29b/0x390
 process_one_work+0x4c5/0xa10
 ? __pfx_process_one_work+0x10/0x10
 ? __list_add_valid_or_report+0x37/0x120
 worker_thread+0x2f1/0x5a0
 ? __kthread_parkme+0xde/0x100
 ? __pfx_worker_thread+0x10/0x10
 kthread+0x1fe/0x380
 ? kthread+0x10f/0x380
 ? __pfx_kthread+0x10/0x10
 ? local_clock_noinstr+0xd/0xd0
 ? ret_from_fork+0x1b/0x1f0
 ? local_clock+0x15/0x30
 ? lock_release+0x29b/0x390
 ? rcu_is_watching+0x20/0x50
 ? __pfx_kthread+0x10/0x10
 ret_from_fork+0x15b/0x1f0
 ? __pfx_kthread+0x10/0x10
 ret_from_fork_asm+0x1a/0x30
 </TASK>
irq event stamp: 1116206
hardirqs last  enabled at (1116205): [<ffffffff8143af42>] __up_console_sem+0x52/0x60
hardirqs last disabled at (1116206): [<ffffffff81399f0e>] queue_delayed_work_on+0x6e/0xc0
softirqs last  enabled at (1116138): [<ffffffffc04562fd>] __smb_send_rqst+0x42d/0x950 [cifs]
softirqs last disabled at (1116136): [<ffffffff823d35e1>] release_sock+0x21/0xf0
---
 fs/smb/client/cifsglob.h |  1 +
 fs/smb/client/smb2pdu.c  | 10 ++++------
 2 files changed, 5 insertions(+), 6 deletions(-)
fs/smb/client/cifsglob.h +1 -0
index 56381cbb63990..b89db44e0a8e3 100644
@@ -1303,6 +1303,7 @@ struct cifs_tcon {
 	bool use_persistent:1; /* use persistent instead of durable handles */
 	bool no_lease:1;    /* Do not request leases on files or directories */
 	bool use_witness:1; /* use witness protocol */
+	bool dummy:1; /* dummy tcon used for reconnecting channels */
 	__le32 capabilities;
 	__u32 share_flags;
 	__u32 maximal_access;
fs/smb/client/smb2pdu.c +4 -6
index 72903265b1706..c3b212175b2bf 100644
@@ -423,9 +423,9 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon,
 		free_xid(xid);
 		ses->flags &= ~CIFS_SES_FLAGS_PENDING_QUERY_INTERFACES;
 
-		/* regardless of rc value, setup polling */
-		queue_delayed_work(cifsiod_wq, &tcon->query_interfaces,
-				   (SMB_INTERFACE_POLL_INTERVAL * HZ));
+		if (!tcon->ipc && !tcon->dummy)
+			queue_delayed_work(cifsiod_wq, &tcon->query_interfaces,
+					   (SMB_INTERFACE_POLL_INTERVAL * HZ));
 
 		mutex_unlock(&ses->session_mutex);
 
@@ -4221,10 +4221,8 @@ void smb2_reconnect_server(struct work_struct *work)
 		}
 		goto done;
 	}
-
 	tcon->status = TID_GOOD;
-	tcon->retry = false;
-	tcon->need_reconnect = false;
+	tcon->dummy = true;
 
 	/* now reconnect sessions for necessary channels */
 	list_for_each_entry_safe(ses, ses2, &tmp_ses_list, rlist) {
Signature · 1 line
2.39.5
author maintainer
#88 lore raw

PATCH 88/178 net: usb: lan78xx: fix WARN in __netif_napi_del_locked on disconnect

In reply to Greg Kroah-Hartman

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

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

From: Oleksij Rempel <[email protected]>

[ Upstream commit 6c7ffc9af7186ed79403a3ffee9a1e5199fc7450 ]

Remove redundant netif_napi_del() call from disconnect path.

A WARN may be triggered in __netif_napi_del_locked() during USB device disconnect:

  WARNING: CPU: 0 PID: 11 at net/core/dev.c:7417 __netif_napi_del_locked+0x2b4/0x350

This happens because netif_napi_del() is called in the disconnect path while NAPI is still enabled. However, it is not necessary to call netif_napi_del() explicitly, since unregister_netdev() will handle NAPI teardown automatically and safely. Removing the redundant call avoids triggering the warning.

Full trace:
lan78xx 1-1:1.0 enu1: Failed to read register index 0x000000c4. ret = -ENODEV lan78xx 1-1:1.0 enu1: Failed to set MAC down with error -ENODEV lan78xx 1-1:1.0 enu1: Link is Down
lan78xx 1-1:1.0 enu1: Failed to read register index 0x00000120. ret = -ENODEV

 ------------[ cut here ]------------
 WARNING: CPU: 0 PID: 11 at net/core/dev.c:7417 __netif_napi_del_locked+0x2b4/0x350
 Modules linked in: flexcan can_dev fuse
 CPU: 0 UID: 0 PID: 11 Comm: kworker/0:1 Not tainted 6.16.0-rc2-00624-ge926949dab03 #9 PREEMPT
 Hardware name: SKOV IMX8MP CPU revC - bd500 (DT)
 Workqueue: usb_hub_wq hub_event
 pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
 pc : __netif_napi_del_locked+0x2b4/0x350
 lr : __netif_napi_del_locked+0x7c/0x350
 sp : ffffffc085b673c0
 x29: ffffffc085b673c0 x28: ffffff800b7f2000 x27: ffffff800b7f20d8
 x26: ffffff80110bcf58 x25: ffffff80110bd978 x24: 1ffffff0022179eb
 x23: ffffff80110bc000 x22: ffffff800b7f5000 x21: ffffff80110bc000
 x20: ffffff80110bcf38 x19: ffffff80110bcf28 x18: dfffffc000000000
 x17: ffffffc081578940 x16: ffffffc08284cee0 x15: 0000000000000028
 x14: 0000000000000006 x13: 0000000000040000 x12: ffffffb0022179e8
 x11: 1ffffff0022179e7 x10: ffffffb0022179e7 x9 : dfffffc000000000
 x8 : 0000004ffdde8619 x7 : ffffff80110bcf3f x6 : 0000000000000001
 x5 : ffffff80110bcf38 x4 : ffffff80110bcf38 x3 : 0000000000000000
 x2 : 0000000000000000 x1 : 1ffffff0022179e7 x0 : 0000000000000000
 Call trace:
  __netif_napi_del_locked+0x2b4/0x350 (P)
  lan78xx_disconnect+0xf4/0x360
  usb_unbind_interface+0x158/0x718
  device_remove+0x100/0x150
  device_release_driver_internal+0x308/0x478
  device_release_driver+0x1c/0x30
  bus_remove_device+0x1a8/0x368
  device_del+0x2e0/0x7b0
  usb_disable_device+0x244/0x540
  usb_disconnect+0x220/0x758
  hub_event+0x105c/0x35e0
  process_one_work+0x760/0x17b0
  worker_thread+0x768/0xce8
  kthread+0x3bc/0x690
  ret_from_fork+0x10/0x20
 irq event stamp: 211604
 hardirqs last  enabled at (211603): [<ffffffc0828cc9ec>] _raw_spin_unlock_irqrestore+0x84/0x98
 hardirqs last disabled at (211604): [<ffffffc0828a9a84>] el1_dbg+0x24/0x80
 softirqs last  enabled at (211296): [<ffffffc080095f10>] handle_softirqs+0x820/0xbc8
 softirqs last disabled at (210993): [<ffffffc080010288>] __do_softirq+0x18/0x20
 ---[ end trace 0000000000000000 ]---
 lan78xx 1-1:1.0 enu1: failed to kill vid 0081/0
---
 drivers/net/usb/lan78xx.c | 2 --
 1 file changed, 2 deletions(-)
drivers/net/usb/lan78xx.c +0 -2
index e4f1663b62047..5f014b8626853 100644
@@ -4344,8 +4344,6 @@ static void lan78xx_disconnect(struct usb_interface *intf)
 	if (!dev)
 		return;
 
-	netif_napi_del(&dev->napi);
-
 	udev = interface_to_usbdev(intf);
 	net = dev->net;
 
Signature · 1 line
2.39.5
author maintainer
#71 lore raw

PATCH 71/178 ethernet: atl1: Add missing DMA mapping error checks and count errors

In reply to Greg Kroah-Hartman

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

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

From: Thomas Fourier <[email protected]>

[ Upstream commit d72411d20905180cdc452c553be17481b24463d2 ]

The dma_map_XXX() functions can fail and must be checked using dma_mapping_error(). This patch adds proper error handling for all DMA mapping calls.

In atl1_alloc_rx_buffers(), if DMA mapping fails, the buffer is deallocated and marked accordingly.

In atl1_tx_map(), previously mapped buffers are unmapped and the packet is dropped on failure.

If atl1_xmit_frame() drops the packet, increment the tx_error counter.

Fixes f3cc28c79760 ("Add Attansic L1 ethernet driver.")
---
 drivers/net/ethernet/atheros/atlx/atl1.c | 79 +++++++++++++++++-------
 1 file changed, 57 insertions(+), 22 deletions(-)
drivers/net/ethernet/atheros/atlx/atl1.c +57 -22
index 38cd84b7677cf..86663aa0de17e 100644
@@ -1861,14 +1861,21 @@ static u16 atl1_alloc_rx_buffers(struct atl1_adapter *adapter)
 			break;
 		}
 
-		buffer_info->alloced = 1;
-		buffer_info->skb = skb;
-		buffer_info->length = (u16) adapter->rx_buffer_len;
 		page = virt_to_page(skb->data);
 		offset = offset_in_page(skb->data);
 		buffer_info->dma = dma_map_page(&pdev->dev, page, offset,
 						adapter->rx_buffer_len,
 						DMA_FROM_DEVICE);
+		if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
+			kfree_skb(skb);
+			adapter->soft_stats.rx_dropped++;
+			break;
+		}
+
+		buffer_info->alloced = 1;
+		buffer_info->skb = skb;
+		buffer_info->length = (u16)adapter->rx_buffer_len;
+
 		rfd_desc->buffer_addr = cpu_to_le64(buffer_info->dma);
 		rfd_desc->buf_len = cpu_to_le16(adapter->rx_buffer_len);
 		rfd_desc->coalese = 0;
@@ -2183,8 +2190,8 @@ static int atl1_tx_csum(struct atl1_adapter *adapter, struct sk_buff *skb,
 	return 0;
 }
 
-static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
-	struct tx_packet_desc *ptpd)
+static bool atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
+			struct tx_packet_desc *ptpd)
 {
 	struct atl1_tpd_ring *tpd_ring = &adapter->tpd_ring;
 	struct atl1_buffer *buffer_info;
@@ -2194,6 +2201,7 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 	unsigned int nr_frags;
 	unsigned int f;
 	int retval;
+	u16 first_mapped;
 	u16 next_to_use;
 	u16 data_len;
 	u8 hdr_len;
@@ -2201,6 +2209,7 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 	buf_len -= skb->data_len;
 	nr_frags = skb_shinfo(skb)->nr_frags;
 	next_to_use = atomic_read(&tpd_ring->next_to_use);
+	first_mapped = next_to_use;
 	buffer_info = &tpd_ring->buffer_info[next_to_use];
 	BUG_ON(buffer_info->skb);
 	/* put skb in last TPD */
@@ -2216,6 +2225,8 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 		buffer_info->dma = dma_map_page(&adapter->pdev->dev, page,
 						offset, hdr_len,
 						DMA_TO_DEVICE);
+		if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma))
+			goto dma_err;
 
 		if (++next_to_use == tpd_ring->count)
 			next_to_use = 0;
@@ -2242,6 +2253,9 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 								page, offset,
 								buffer_info->length,
 								DMA_TO_DEVICE);
+				if (dma_mapping_error(&adapter->pdev->dev,
+						      buffer_info->dma))
+					goto dma_err;
 				if (++next_to_use == tpd_ring->count)
 					next_to_use = 0;
 			}
@@ -2254,6 +2268,8 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 		buffer_info->dma = dma_map_page(&adapter->pdev->dev, page,
 						offset, buf_len,
 						DMA_TO_DEVICE);
+		if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma))
+			goto dma_err;
 		if (++next_to_use == tpd_ring->count)
 			next_to_use = 0;
 	}
@@ -2277,6 +2293,9 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 			buffer_info->dma = skb_frag_dma_map(&adapter->pdev->dev,
 				frag, i * ATL1_MAX_TX_BUF_LEN,
 				buffer_info->length, DMA_TO_DEVICE);
+			if (dma_mapping_error(&adapter->pdev->dev,
+					      buffer_info->dma))
+				goto dma_err;
 
 			if (++next_to_use == tpd_ring->count)
 				next_to_use = 0;
@@ -2285,6 +2304,22 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb,
 
 	/* last tpd's buffer-info */
 	buffer_info->skb = skb;
+
+	return true;
+
+ dma_err:
+	while (first_mapped != next_to_use) {
+		buffer_info = &tpd_ring->buffer_info[first_mapped];
+		dma_unmap_page(&adapter->pdev->dev,
+			       buffer_info->dma,
+			       buffer_info->length,
+			       DMA_TO_DEVICE);
+		buffer_info->dma = 0;
+
+		if (++first_mapped == tpd_ring->count)
+			first_mapped = 0;
+	}
+	return false;
 }
 
 static void atl1_tx_queue(struct atl1_adapter *adapter, u16 count,
@@ -2355,10 +2390,8 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb,
 
 	len = skb_headlen(skb);
 
-	if (unlikely(skb->len <= 0)) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
+	if (unlikely(skb->len <= 0))
+		goto drop_packet;
 
 	nr_frags = skb_shinfo(skb)->nr_frags;
 	for (f = 0; f < nr_frags; f++) {
@@ -2371,10 +2404,9 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb,
 	if (mss) {
 		if (skb->protocol == htons(ETH_P_IP)) {
 			proto_hdr_len = skb_tcp_all_headers(skb);
-			if (unlikely(proto_hdr_len > len)) {
-				dev_kfree_skb_any(skb);
-				return NETDEV_TX_OK;
-			}
+			if (unlikely(proto_hdr_len > len))
+				goto drop_packet;
+
 			/* need additional TPD ? */
 			if (proto_hdr_len != len)
 				count += (len - proto_hdr_len +
@@ -2406,23 +2438,26 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb,
 	}
 
 	tso = atl1_tso(adapter, skb, ptpd);
-	if (tso < 0) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
+	if (tso < 0)
+		goto drop_packet;
 
 	if (!tso) {
 		ret_val = atl1_tx_csum(adapter, skb, ptpd);
-		if (ret_val < 0) {
-			dev_kfree_skb_any(skb);
-			return NETDEV_TX_OK;
-		}
+		if (ret_val < 0)
+			goto drop_packet;
 	}
 
-	atl1_tx_map(adapter, skb, ptpd);
+	if (!atl1_tx_map(adapter, skb, ptpd))
+		goto drop_packet;
+
 	atl1_tx_queue(adapter, count, ptpd);
 	atl1_update_mailbox(adapter);
 	return NETDEV_TX_OK;
+
+drop_packet:
+	adapter->soft_stats.tx_errors++;
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_OK;
 }
 
 static int atl1_rings_clean(struct napi_struct *napi, int budget)
Signature · 1 line
2.39.5
author maintainer
#89 lore raw

PATCH 89/178 drm/vmwgfx: Fix guests running with TDX/SEV

In reply to Greg Kroah-Hartman

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

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

From: Marko Kiiskila <[email protected]>

[ Upstream commit 7dfede7d7edd18c0c91ca854cde8eaaf4ccf97ea ]

Commit 81256a50aa0f ("x86/mm: Make memremap(MEMREMAP_WB) map memory as encrypted by default") changed the default behavior of memremap(MEMREMAP_WB) and started mapping memory as encrypted. The driver requires the fifo memory to be decrypted to communicate with the host but was relaying on the old default behavior of memremap(MEMREMAP_WB) and thus broke.

Fix it by explicitly specifying the desired behavior and passing MEMREMAP_DEC to memremap.

Fixes 81256a50aa0f ("x86/mm: Make memremap(MEMREMAP_WB) map memory as encrypted by default")
Cc Kirill A. Shutemov <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Andrew Morton <[email protected]>
Cc Dave Hansen <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Peter Zijlstra <[email protected]>
---
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +1 -1
index 0f32471c85332..55c822a61b9ad 100644
@@ -769,7 +769,7 @@ static int vmw_setup_pci_resources(struct vmw_private *dev,
 		dev->fifo_mem = devm_memremap(dev->drm.dev,
 					      fifo_start,
 					      fifo_size,
-					      MEMREMAP_WB);
+					      MEMREMAP_WB | MEMREMAP_DEC);
 
 		if (IS_ERR(dev->fifo_mem)) {
 			drm_err(&dev->drm,
Signature · 1 line
2.39.5
author maintainer
#90 lore raw

PATCH 90/178 drm/i915/gt: Fix timeline left held on VMA alloc error

In reply to Greg Kroah-Hartman

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

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

From: Janusz Krzysztofik <[email protected]>

[ Upstream commit a5aa7bc1fca78c7fa127d9e33aa94a0c9066c1d6 ]

The following error has been reported sporadically by CI when a test unbinds the i915 driver on a ring submission platform:

<4> [239.330153] ------------[ cut here ]------------
<4> [239.330166] i915 0000:00:02.0: [drm] drm_WARN_ON(dev_priv->mm.shrink_count)
<4> [239.330196] WARNING: CPU: 1 PID: 18570 at drivers/gpu/drm/i915/i915_gem.c:1309 i915_gem_cleanup_early+0x13e/0x150 [i915]
...
<4> [239.330640] RIP: 0010:i915_gem_cleanup_early+0x13e/0x150 [i915]
...
<4> [239.330942] Call Trace:

<4> [239.330944]  <TASK>
<4> [239.330949]  i915_driver_late_release+0x2b/0xa0 [i915]
<4> [239.331202]  i915_driver_release+0x86/0xa0 [i915]
<4> [239.331482]  devm_drm_dev_init_release+0x61/0x90
<4> [239.331494]  devm_action_release+0x15/0x30
<4> [239.331504]  release_nodes+0x3d/0x120
<4> [239.331517]  devres_release_all+0x96/0xd0
<4> [239.331533]  device_unbind_cleanup+0x12/0x80
<4> [239.331543]  device_release_driver_internal+0x23a/0x280
<4> [239.331550]  ? bus_find_device+0xa5/0xe0
<4> [239.331563]  device_driver_detach+0x14/0x20

...
<4> [357.719679] ---[ end trace 0000000000000000 ]---

If the test also unloads the i915 module then that's followed with:

<3> [357.787478] =============================================================================
<3> [357.788006] BUG i915_vma (Tainted: G U W N ): Objects remaining on __kmem_cache_shutdown()
<3> [357.788031] ----------------------------------------------------------------------------- <3> [357.788204] Object 0xffff888109e7f480 @offset=29824
<3> [357.788670] Allocated in i915_vma_instance+0xee/0xc10 [i915] age=292729 cpu=4 pid=2244

<4> [357.788994]  i915_vma_instance+0xee/0xc10 [i915]
<4> [357.789290]  init_status_page+0x7b/0x420 [i915]
<4> [357.789532]  intel_engines_init+0x1d8/0x980 [i915]
<4> [357.789772]  intel_gt_init+0x175/0x450 [i915]
<4> [357.790014]  i915_gem_init+0x113/0x340 [i915]
<4> [357.790281]  i915_driver_probe+0x847/0xed0 [i915]
<4> [357.790504]  i915_pci_probe+0xe6/0x220 [i915]

...

Closer analysis of CI results history has revealed a dependency of the error on a few IGT tests, namely:
- igt@api_intel_allocator@fork-simple-stress-signal,
- igt@api_intel_allocator@two-level-inception-interruptible,
- igt@gem_linear_blits@interruptible,
- igt@prime_mmap_coherency@ioctl-errors,
which invisibly trigger the issue, then exhibited with first driver unbind attempt.

All of the above tests perform actions which are actively interrupted with signals. Further debugging has allowed to narrow that scope down to DRM_IOCTL_I915_GEM_EXECBUFFER2, and ring_context_alloc(), specific to ring submission, in particular.

If successful then that function, or its execlists or GuC submission equivalent, is supposed to be called only once per GEM context engine, followed by raise of a flag that prevents the function from being called again. The function is expected to unwind its internal errors itself, so it may be safely called once more after it returns an error.

In case of ring submission, the function first gets a reference to the engine's legacy timeline and then allocates a VMA. If the VMA allocation fails, e.g. when i915_vma_instance() called from inside is interrupted with a signal, then ring_context_alloc() fails, leaving the timeline held referenced. On next I915_GEM_EXECBUFFER2 IOCTL, another reference to the timeline is got, and only that last one is put on successful completion. As a consequence, the legacy timeline, with its underlying engine status page's VMA object, is still held and not released on driver unbind.

Get the legacy timeline only after successful allocation of the context engine's VMA.

v2: Add a note on other submission methods (Krzysztof Karas):
Both execlists and GuC submission use lrc_alloc() which seems free from a similar issue.

Fixes 75d0a7f31eec ("drm/i915: Lift timeline into intel_context")
Cc Chris Wilson <[email protected]>
Cc Matthew Auld <[email protected]>
Cc Krzysztof Karas <[email protected]>
(cherry picked from commit cc43422b3cc79eacff4c5a8ba0d224688ca9dd4f)
---
 drivers/gpu/drm/i915/gt/intel_ring_submission.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/gt/intel_ring_submission.c +2 -1
index 6e9977b2d1802..44a4b6723582f 100644
@@ -604,7 +604,6 @@ static int ring_context_alloc(struct intel_context *ce)
 	/* One ringbuffer to rule them all */
 	GEM_BUG_ON(!engine->legacy.ring);
 	ce->ring = engine->legacy.ring;
-	ce->timeline = intel_timeline_get(engine->legacy.timeline);
 
 	GEM_BUG_ON(ce->state);
 	if (engine->context_size) {
@@ -617,6 +616,8 @@ static int ring_context_alloc(struct intel_context *ce)
 		ce->state = vma;
 	}
 
+	ce->timeline = intel_timeline_get(engine->legacy.timeline);
+
 	return 0;
 }
 
Signature · 1 line
2.39.5
author maintainer
#91 lore raw

PATCH 91/178 drm/i915/gsc: mei interrupt top half should be in irq disabled context

In reply to Greg Kroah-Hartman

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

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

From: Junxiao Chang <[email protected]>

[ Upstream commit 8cadce97bf264ed478669c6f32d5603b34608335 ]

MEI GSC interrupt comes from i915. It has top half and bottom half. Top half is called from i915 interrupt handler. It should be in irq disabled context.

With RT kernel, by default i915 IRQ handler is in threaded IRQ. MEI GSC top half might be in threaded IRQ context. generic_handle_irq_safe API could be called from either IRQ or process context, it disables local IRQ then calls MEI GSC interrupt top half.

This change fixes A380/A770 GPU boot hang issue with RT kernel.

Fixes 1e3dc1d8622b ("drm/i915/gsc: add gsc as a mei auxiliary device")
(cherry picked from commit dccf655f69002d496a527ba441b4f008aa5bebbf)
---
 drivers/gpu/drm/i915/gt/intel_gsc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/i915/gt/intel_gsc.c +1 -1
index 1e925c75fb080..c43febc862dc3 100644
@@ -284,7 +284,7 @@ static void gsc_irq_handler(struct intel_gt *gt, unsigned int intf_id)
 	if (gt->gsc.intf[intf_id].irq < 0)
 		return;
 
-	ret = generic_handle_irq(gt->gsc.intf[intf_id].irq);
+	ret = generic_handle_irq_safe(gt->gsc.intf[intf_id].irq);
 	if (ret)
 		gt_err_ratelimited(gt, "error handling GSC irq: %d\n", ret);
 }
Signature · 1 line
2.39.5
author maintainer
#92 lore raw

PATCH 92/178 nvme-multipath: fix suspicious RCU usage warning

In reply to Greg Kroah-Hartman

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

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

From: Geliang Tang <[email protected]>

[ Upstream commit d6811074203b13f715ce2480ac64c5b1c773f2a5 ]

When I run the NVME over TCP test in virtme-ng, I get the following "suspicious RCU usage" warning in nvme_mpath_add_sysfs_link():

'''

[    5.024557][   T44] nvmet: Created nvm controller 1 for subsystem nqn.2025-06.org.nvmexpress.mptcp for NQN nqn.2014-08.org.nvmexpress:uuid:f7f6b5e0-ff97-4894-98ac-c85309e0bc77.
[    5.027401][  T183] nvme nvme0: creating 2 I/O queues.
[    5.029017][  T183] nvme nvme0: mapped 2/0/0 default/read/poll queues.
[    5.032587][  T183] nvme nvme0: new ctrl: NQN "nqn.2025-06.org.nvmexpress.mptcp", addr 127.0.0.1:4420, hostnqn: nqn.2014-08.org.nvmexpress:uuid:f7f6b5e0-ff97-4894-98ac-c85309e0bc77
[    5.042214][   T25]
[    5.042440][   T25] =============================
[    5.042579][   T25] WARNING: suspicious RCU usage
[    5.042705][   T25] 6.16.0-rc3+ #23 Not tainted
[    5.042812][   T25] -----------------------------
[    5.042934][   T25] drivers/nvme/host/multipath.c:1203 RCU-list traversed in non-reader section!!
[    5.043111][   T25]
[    5.043111][   T25] other info that might help us debug this:
[    5.043111][   T25]
[    5.043341][   T25]
[    5.043341][   T25] rcu_scheduler_active = 2, debug_locks = 1
[    5.043502][   T25] 3 locks held by kworker/u9:0/25:
[    5.043615][   T25]  #0: ffff888008730948 ((wq_completion)async){+.+.}-{0:0}, at: process_one_work+0x7ed/0x1350
[    5.043830][   T25]  #1: ffffc900001afd40 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0xcf3/0x1350
[    5.044084][   T25]  #2: ffff888013ee0020 (&head->srcu){.+.+}-{0:0}, at: nvme_mpath_add_sysfs_link.part.0+0xb4/0x3a0
[    5.044300][   T25]
[    5.044300][   T25] stack backtrace:
[    5.044439][   T25] CPU: 0 UID: 0 PID: 25 Comm: kworker/u9:0 Not tainted 6.16.0-rc3+ #23 PREEMPT(full)
[    5.044441][   T25] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[    5.044442][   T25] Workqueue: async async_run_entry_fn
[    5.044445][   T25] Call Trace:
[    5.044446][   T25]  <TASK>
[    5.044449][   T25]  dump_stack_lvl+0x6f/0xb0
[    5.044453][   T25]  lockdep_rcu_suspicious.cold+0x4f/0xb1
[    5.044457][   T25]  nvme_mpath_add_sysfs_link.part.0+0x2fb/0x3a0
[    5.044459][   T25]  ? queue_work_on+0x90/0xf0
[    5.044461][   T25]  ? lockdep_hardirqs_on+0x78/0x110
[    5.044466][   T25]  nvme_mpath_set_live+0x1e9/0x4f0
[    5.044470][   T25]  nvme_mpath_add_disk+0x240/0x2f0
[    5.044472][   T25]  ? __pfx_nvme_mpath_add_disk+0x10/0x10
[    5.044475][   T25]  ? add_disk_fwnode+0x361/0x580
[    5.044480][   T25]  nvme_alloc_ns+0x81c/0x17c0
[    5.044483][   T25]  ? kasan_quarantine_put+0x104/0x240
[    5.044487][   T25]  ? __pfx_nvme_alloc_ns+0x10/0x10
[    5.044495][   T25]  ? __pfx_nvme_find_get_ns+0x10/0x10
[    5.044496][   T25]  ? rcu_read_lock_any_held+0x45/0xa0
[    5.044498][   T25]  ? validate_chain+0x232/0x4f0
[    5.044503][   T25]  nvme_scan_ns+0x4c8/0x810
[    5.044506][   T25]  ? __pfx_nvme_scan_ns+0x10/0x10
[    5.044508][   T25]  ? find_held_lock+0x2b/0x80
[    5.044512][   T25]  ? ktime_get+0x16d/0x220
[    5.044517][   T25]  ? kvm_clock_get_cycles+0x18/0x30
[    5.044520][   T25]  ? __pfx_nvme_scan_ns_async+0x10/0x10
[    5.044522][   T25]  async_run_entry_fn+0x97/0x560
[    5.044523][   T25]  ? rcu_is_watching+0x12/0xc0
[    5.044526][   T25]  process_one_work+0xd3c/0x1350
[    5.044532][   T25]  ? __pfx_process_one_work+0x10/0x10
[    5.044536][   T25]  ? assign_work+0x16c/0x240
[    5.044539][   T25]  worker_thread+0x4da/0xd50
[    5.044545][   T25]  ? __pfx_worker_thread+0x10/0x10
[    5.044546][   T25]  kthread+0x356/0x5c0
[    5.044548][   T25]  ? __pfx_kthread+0x10/0x10
[    5.044549][   T25]  ? ret_from_fork+0x1b/0x2e0
[    5.044552][   T25]  ? __lock_release.isra.0+0x5d/0x180
[    5.044553][   T25]  ? ret_from_fork+0x1b/0x2e0
[    5.044555][   T25]  ? rcu_is_watching+0x12/0xc0
[    5.044557][   T25]  ? __pfx_kthread+0x10/0x10
[    5.044559][   T25]  ret_from_fork+0x218/0x2e0
[    5.044561][   T25]  ? __pfx_kthread+0x10/0x10
[    5.044562][   T25]  ret_from_fork_asm+0x1a/0x30
[    5.044570][   T25]  </TASK>

'''

This patch uses sleepable RCU version of helper list_for_each_entry_srcu() instead of list_for_each_entry_rcu() to fix it.

Fixes 4dbd2b2ebe4c ("nvme-multipath: Add visibility for round-robin io-policy")
---
 drivers/nvme/host/multipath.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/nvme/host/multipath.c +2 -1
index cf0ef4745564c..700dfbd5a4512 100644
@@ -1050,7 +1050,8 @@ void nvme_mpath_add_sysfs_link(struct nvme_ns_head *head)
 	 */
 	srcu_idx = srcu_read_lock(&head->srcu);
 
-	list_for_each_entry_rcu(ns, &head->list, siblings) {
+	list_for_each_entry_srcu(ns, &head->list, siblings,
+				 srcu_read_lock_held(&head->srcu)) {
 		/*
 		 * Ensure that ns path disk node is already added otherwise we
 		 * may get invalid kobj name for target
Signature · 1 line
2.39.5
author maintainer
#93 lore raw

PATCH 93/178 idpf: return 0 size for RSS key if not supported

In reply to Greg Kroah-Hartman

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

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

From: Michal Swiatkowski <[email protected]>

[ Upstream commit f77bf1ebf8ff6301ccdbc346f7b52db928f9cbf8 ]

Returning -EOPNOTSUPP from function returning u32 is leading to cast and invalid size value as a result.

-EOPNOTSUPP as a size probably will lead to allocation fail.

Command: ethtool -x eth0
It is visible on all devices that don't have RSS caps set.

[  136.615917] Call Trace:
[  136.615921]  <TASK>
[  136.615927]  ? __warn+0x89/0x130
[  136.615942]  ? __alloc_frozen_pages_noprof+0x322/0x330
[  136.615953]  ? report_bug+0x164/0x190
[  136.615968]  ? handle_bug+0x58/0x90
[  136.615979]  ? exc_invalid_op+0x17/0x70
[  136.615987]  ? asm_exc_invalid_op+0x1a/0x20
[  136.616001]  ? rss_prepare_get.constprop.0+0xb9/0x170
[  136.616016]  ? __alloc_frozen_pages_noprof+0x322/0x330
[  136.616028]  __alloc_pages_noprof+0xe/0x20
[  136.616038]  ___kmalloc_large_node+0x80/0x110
[  136.616072]  __kmalloc_large_node_noprof+0x1d/0xa0
[  136.616081]  __kmalloc_noprof+0x32c/0x4c0
[  136.616098]  ? rss_prepare_get.constprop.0+0xb9/0x170
[  136.616105]  rss_prepare_get.constprop.0+0xb9/0x170
[  136.616114]  ethnl_default_doit+0x107/0x3d0
[  136.616131]  genl_family_rcv_msg_doit+0x100/0x160
[  136.616147]  genl_rcv_msg+0x1b8/0x2c0
[  136.616156]  ? __pfx_ethnl_default_doit+0x10/0x10
[  136.616168]  ? __pfx_genl_rcv_msg+0x10/0x10
[  136.616176]  netlink_rcv_skb+0x58/0x110
[  136.616186]  genl_rcv+0x28/0x40
[  136.616195]  netlink_unicast+0x19b/0x290
[  136.616206]  netlink_sendmsg+0x222/0x490
[  136.616215]  __sys_sendto+0x1fd/0x210
[  136.616233]  __x64_sys_sendto+0x24/0x30
[  136.616242]  do_syscall_64+0x82/0x160
[  136.616252]  ? __sys_recvmsg+0x83/0xe0
[  136.616265]  ? syscall_exit_to_user_mode+0x10/0x210
[  136.616275]  ? do_syscall_64+0x8e/0x160
[  136.616282]  ? __count_memcg_events+0xa1/0x130
[  136.616295]  ? count_memcg_events.constprop.0+0x1a/0x30
[  136.616306]  ? handle_mm_fault+0xae/0x2d0
[  136.616319]  ? do_user_addr_fault+0x379/0x670
[  136.616328]  ? clear_bhb_loop+0x45/0xa0
[  136.616340]  ? clear_bhb_loop+0x45/0xa0
[  136.616349]  ? clear_bhb_loop+0x45/0xa0
[  136.616359]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  136.616369] RIP: 0033:0x7fd30ba7b047
[  136.616376] Code: 0c 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f 1e fa 80 3d bd d5 0c 00 00 41 89 ca 74 10 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 71 c3 55 48 83 ec 30 44 89 4c 24 2c 4c 89 44
[  136.616381] RSP: 002b:00007ffde1796d68 EFLAGS: 00000202 ORIG_RAX: 000000000000002c
[  136.616388] RAX: ffffffffffffffda RBX: 000055d7bd89f2a0 RCX: 00007fd30ba7b047
[  136.616392] RDX: 0000000000000028 RSI: 000055d7bd89f3b0 RDI: 0000000000000003
[  136.616396] RBP: 00007ffde1796e10 R08: 00007fd30bb4e200 R09: 000000000000000c
[  136.616399] R10: 0000000000000000 R11: 0000000000000202 R12: 000055d7bd89f340
[  136.616403] R13: 000055d7bd89f3b0 R14: 000055d78943f200 R15: 0000000000000000
---
 drivers/net/ethernet/intel/idpf/idpf_ethtool.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/idpf/idpf_ethtool.c +2 -2
index 59b1a1a099967..f72420cf68216 100644
@@ -46,7 +46,7 @@ static u32 idpf_get_rxfh_key_size(struct net_device *netdev)
 	struct idpf_vport_user_config_data *user_config;
 
 	if (!idpf_is_cap_ena_all(np->adapter, IDPF_RSS_CAPS, IDPF_CAP_RSS))
-		return -EOPNOTSUPP;
+		return 0;
 
 	user_config = &np->adapter->vport_config[np->vport_idx]->user_config;
 
@@ -65,7 +65,7 @@ static u32 idpf_get_rxfh_indir_size(struct net_device *netdev)
 	struct idpf_vport_user_config_data *user_config;
 
 	if (!idpf_is_cap_ena_all(np->adapter, IDPF_RSS_CAPS, IDPF_CAP_RSS))
-		return -EOPNOTSUPP;
+		return 0;
 
 	user_config = &np->adapter->vport_config[np->vport_idx]->user_config;
 
Signature · 1 line
2.39.5
author maintainer
#94 lore raw

PATCH 94/178 idpf: convert control queue mutex to a spinlock

In reply to Greg Kroah-Hartman

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

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

From: Ahmed Zaki <[email protected]>

[ Upstream commit b2beb5bb2cd90d7939e470ed4da468683f41baa3 ]

With VIRTCHNL2_CAP_MACFILTER enabled, the following warning is generated on module load:

[  324.701677] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:578
[  324.701684] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 1582, name: NetworkManager
[  324.701689] preempt_count: 201, expected: 0
[  324.701693] RCU nest depth: 0, expected: 0
[  324.701697] 2 locks held by NetworkManager/1582:
[  324.701702]  #0: ffffffff9f7be770 (rtnl_mutex){....}-{3:3}, at: rtnl_newlink+0x791/0x21e0
[  324.701730]  #1: ff1100216c380368 (_xmit_ETHER){....}-{2:2}, at: __dev_open+0x3f0/0x870
[  324.701749] Preemption disabled at:
[  324.701752] [<ffffffff9cd23b9d>] __dev_open+0x3dd/0x870
[  324.701765] CPU: 30 UID: 0 PID: 1582 Comm: NetworkManager Not tainted 6.15.0-rc5+ #2 PREEMPT(voluntary)
[  324.701771] Hardware name: Intel Corporation M50FCP2SBSTD/M50FCP2SBSTD, BIOS SE5C741.86B.01.01.0001.2211140926 11/14/2022
[  324.701774] Call Trace:
[  324.701777]  <TASK>
[  324.701779]  dump_stack_lvl+0x5d/0x80
[  324.701788]  ? __dev_open+0x3dd/0x870
[  324.701793]  __might_resched.cold+0x1ef/0x23d
<..>
[  324.701818]  __mutex_lock+0x113/0x1b80
<..>
[  324.701917]  idpf_ctlq_clean_sq+0xad/0x4b0 [idpf]
[  324.701935]  ? kasan_save_track+0x14/0x30
[  324.701941]  idpf_mb_clean+0x143/0x380 [idpf]
<..>
[  324.701991]  idpf_send_mb_msg+0x111/0x720 [idpf]
[  324.702009]  idpf_vc_xn_exec+0x4cc/0x990 [idpf]
[  324.702021]  ? rcu_is_watching+0x12/0xc0
[  324.702035]  idpf_add_del_mac_filters+0x3ed/0xb50 [idpf]
<..>
[  324.702122]  __hw_addr_sync_dev+0x1cf/0x300
[  324.702126]  ? find_held_lock+0x32/0x90
[  324.702134]  idpf_set_rx_mode+0x317/0x390 [idpf]
[  324.702152]  __dev_open+0x3f8/0x870
[  324.702159]  ? __pfx___dev_open+0x10/0x10
[  324.702174]  __dev_change_flags+0x443/0x650
<..>
[  324.702208]  netif_change_flags+0x80/0x160
[  324.702218]  do_setlink.isra.0+0x16a0/0x3960
<..>
[  324.702349]  rtnl_newlink+0x12fd/0x21e0

The sequence is as follows:

	rtnl_newlink()->
	__dev_change_flags()->
	__dev_open()->
	dev_set_rx_mode() - >  # disables BH and grabs "dev->addr_list_lock"
	idpf_set_rx_mode() ->  # proceed only if VIRTCHNL2_CAP_MACFILTER is ON
	__dev_uc_sync() ->
	idpf_add_mac_filter ->
	idpf_add_del_mac_filters ->
	idpf_send_mb_msg() ->
	idpf_mb_clean() ->
	idpf_ctlq_clean_sq()   # mutex_lock(cq_lock)

Fix by converting cq_lock to a spinlock. All operations under the new lock are safe except freeing the DMA memory, which may use vunmap(). Fix by requesting a contiguous physical memory for the DMA mapping.

Fixes a251eee62133 ("idpf: add SRIOV support and other ndo_ops")
---
 .../net/ethernet/intel/idpf/idpf_controlq.c   | 23 +++++++++----------
 .../ethernet/intel/idpf/idpf_controlq_api.h   |  2 +-
 drivers/net/ethernet/intel/idpf/idpf_lib.c    | 12 ++++++----
 3 files changed, 20 insertions(+), 17 deletions(-)
drivers/net/ethernet/intel/idpf/idpf_controlq.c +11 -12
index b28991dd18703..48b8e184f3db6 100644
@@ -96,7 +96,7 @@ static void idpf_ctlq_init_rxq_bufs(struct idpf_ctlq_info *cq)
  */
 static void idpf_ctlq_shutdown(struct idpf_hw *hw, struct idpf_ctlq_info *cq)
 {
-	mutex_lock(&cq->cq_lock);
+	spin_lock(&cq->cq_lock);
 
 	/* free ring buffers and the ring itself */
 	idpf_ctlq_dealloc_ring_res(hw, cq);
@@ -104,8 +104,7 @@ static void idpf_ctlq_shutdown(struct idpf_hw *hw, struct idpf_ctlq_info *cq)
 	/* Set ring_size to 0 to indicate uninitialized queue */
 	cq->ring_size = 0;
 
-	mutex_unlock(&cq->cq_lock);
-	mutex_destroy(&cq->cq_lock);
+	spin_unlock(&cq->cq_lock);
 }
 
 /**
@@ -173,7 +172,7 @@ int idpf_ctlq_add(struct idpf_hw *hw,
 
 	idpf_ctlq_init_regs(hw, cq, is_rxq);
 
-	mutex_init(&cq->cq_lock);
+	spin_lock_init(&cq->cq_lock);
 
 	list_add(&cq->cq_list, &hw->cq_list_head);
 
@@ -272,7 +271,7 @@ int idpf_ctlq_send(struct idpf_hw *hw, struct idpf_ctlq_info *cq,
 	int err = 0;
 	int i;
 
-	mutex_lock(&cq->cq_lock);
+	spin_lock(&cq->cq_lock);
 
 	/* Ensure there are enough descriptors to send all messages */
 	num_desc_avail = IDPF_CTLQ_DESC_UNUSED(cq);
@@ -332,7 +331,7 @@ int idpf_ctlq_send(struct idpf_hw *hw, struct idpf_ctlq_info *cq,
 	wr32(hw, cq->reg.tail, cq->next_to_use);
 
 err_unlock:
-	mutex_unlock(&cq->cq_lock);
+	spin_unlock(&cq->cq_lock);
 
 	return err;
 }
@@ -364,7 +363,7 @@ int idpf_ctlq_clean_sq(struct idpf_ctlq_info *cq, u16 *clean_count,
 	if (*clean_count > cq->ring_size)
 		return -EBADR;
 
-	mutex_lock(&cq->cq_lock);
+	spin_lock(&cq->cq_lock);
 
 	ntc = cq->next_to_clean;
 
@@ -397,7 +396,7 @@ int idpf_ctlq_clean_sq(struct idpf_ctlq_info *cq, u16 *clean_count,
 
 	cq->next_to_clean = ntc;
 
-	mutex_unlock(&cq->cq_lock);
+	spin_unlock(&cq->cq_lock);
 
 	/* Return number of descriptors actually cleaned */
 	*clean_count = i;
@@ -435,7 +434,7 @@ int idpf_ctlq_post_rx_buffs(struct idpf_hw *hw, struct idpf_ctlq_info *cq,
 	if (*buff_count > 0)
 		buffs_avail = true;
 
-	mutex_lock(&cq->cq_lock);
+	spin_lock(&cq->cq_lock);
 
 	if (tbp >= cq->ring_size)
 		tbp = 0;
@@ -524,7 +523,7 @@ int idpf_ctlq_post_rx_buffs(struct idpf_hw *hw, struct idpf_ctlq_info *cq,
 		wr32(hw, cq->reg.tail, cq->next_to_post);
 	}
 
-	mutex_unlock(&cq->cq_lock);
+	spin_unlock(&cq->cq_lock);
 
 	/* return the number of buffers that were not posted */
 	*buff_count = *buff_count - i;
@@ -552,7 +551,7 @@ int idpf_ctlq_recv(struct idpf_ctlq_info *cq, u16 *num_q_msg,
 	u16 i;
 
 	/* take the lock before we start messing with the ring */
-	mutex_lock(&cq->cq_lock);
+	spin_lock(&cq->cq_lock);
 
 	ntc = cq->next_to_clean;
 
@@ -614,7 +613,7 @@ int idpf_ctlq_recv(struct idpf_ctlq_info *cq, u16 *num_q_msg,
 
 	cq->next_to_clean = ntc;
 
-	mutex_unlock(&cq->cq_lock);
+	spin_unlock(&cq->cq_lock);
 
 	*num_q_msg = i;
 	if (*num_q_msg == 0)
drivers/net/ethernet/intel/idpf/idpf_controlq_api.h +1 -1
index e8e046ef2f0d7..5890d8adca4a8 100644
@@ -99,7 +99,7 @@ struct idpf_ctlq_info {
 
 	enum idpf_ctlq_type cq_type;
 	int q_id;
-	struct mutex cq_lock;		/* control queue lock */
+	spinlock_t cq_lock;		/* control queue lock */
 	/* used for interrupt processing */
 	u16 next_to_use;
 	u16 next_to_clean;
drivers/net/ethernet/intel/idpf/idpf_lib.c +8 -4
index 2ed801398971c..fe96e20573660 100644
@@ -2329,8 +2329,12 @@ void *idpf_alloc_dma_mem(struct idpf_hw *hw, struct idpf_dma_mem *mem, u64 size)
 	struct idpf_adapter *adapter = hw->back;
 	size_t sz = ALIGN(size, 4096);
 
-	mem->va = dma_alloc_coherent(&adapter->pdev->dev, sz,
-				     &mem->pa, GFP_KERNEL);
+	/* The control queue resources are freed under a spinlock, contiguous
+	 * pages will avoid IOMMU remapping and the use vmap (and vunmap in
+	 * dma_free_*() path.
+	 */
+	mem->va = dma_alloc_attrs(&adapter->pdev->dev, sz, &mem->pa,
+				  GFP_KERNEL, DMA_ATTR_FORCE_CONTIGUOUS);
 	mem->size = sz;
 
 	return mem->va;
@@ -2345,8 +2349,8 @@ void idpf_free_dma_mem(struct idpf_hw *hw, struct idpf_dma_mem *mem)
 {
 	struct idpf_adapter *adapter = hw->back;
 
-	dma_free_coherent(&adapter->pdev->dev, mem->size,
-			  mem->va, mem->pa);
+	dma_free_attrs(&adapter->pdev->dev, mem->size,
+		       mem->va, mem->pa, DMA_ATTR_FORCE_CONTIGUOUS);
 	mem->size = 0;
 	mem->va = NULL;
 	mem->pa = 0;
Signature · 1 line
2.39.5
author maintainer
#95 lore raw

PATCH 95/178 igc: disable L1.2 PCI-E link substate to avoid performance issue

In reply to Greg Kroah-Hartman

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

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

From: Vitaly Lifshits <[email protected]>

[ Upstream commit 0325143b59c6c6d79987afc57d2456e7a20d13b7 ]

I226 devices advertise support for the PCI-E link L1.2 substate. However, due to a hardware limitation, the exit latency from this low-power state is longer than the packet buffer can tolerate under high traffic conditions. This can lead to packet loss and degraded performance.

To mitigate this, disable the L1.2 substate. The increased power draw between L1.1 and L1.2 is insignificant.

---
 drivers/net/ethernet/intel/igc/igc_main.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/net/ethernet/intel/igc/igc_main.c +10 -0
index b1669d7cf4359..7d8cc783b2282 100644
@@ -7046,6 +7046,10 @@ static int igc_probe(struct pci_dev *pdev,
 	adapter->port_num = hw->bus.func;
 	adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
 
+	/* Disable ASPM L1.2 on I226 devices to avoid packet loss */
+	if (igc_is_device_id_i226(hw))
+		pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_2);
+
 	err = pci_save_state(pdev);
 	if (err)
 		goto err_ioremap;
@@ -7426,6 +7430,9 @@ static int __igc_resume(struct device *dev, bool rpm)
 	pci_enable_wake(pdev, PCI_D3hot, 0);
 	pci_enable_wake(pdev, PCI_D3cold, 0);
 
+	if (igc_is_device_id_i226(hw))
+		pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_2);
+
 	if (igc_init_interrupt_scheme(adapter, true)) {
 		netdev_err(netdev, "Unable to allocate memory for queues\n");
 		return -ENOMEM;
@@ -7551,6 +7558,9 @@ static pci_ers_result_t igc_io_slot_reset(struct pci_dev *pdev)
 		pci_enable_wake(pdev, PCI_D3hot, 0);
 		pci_enable_wake(pdev, PCI_D3cold, 0);
 
+		if (igc_is_device_id_i226(hw))
+			pci_disable_link_state_locked(pdev, PCIE_LINK_STATE_L1_2);
+
 		/* In case of PCI error, adapter loses its HW address
 		 * so we should re-assign it here.
 		 */
Signature · 1 line
2.39.5
author maintainer
#96 lore raw

PATCH 96/178 drm/xe: Fix out-of-bounds field write in MI_STORE_DATA_IMM

In reply to Greg Kroah-Hartman

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

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

From: Jia Yao <[email protected]>

[ Upstream commit 2d5cff2b4bc567dcaad7ab5b46c973ba534cc062 ]

According to Bspec, bits 0~9 of MI_STORE_DATA_IMM must not exceed 0x3FE. The macro MI_SDI_NUM_QW(x) evaluates to 2 * x + 1, which means the condition 2 * x + 1 <= 0x3FE must be satisfied. Therefore, the maximum valid value for x is 0x1FE, not 0x1FF.

v2
- Replace 0x1fe with macro MAX_PTE_PER_SDI (Auld, Matthew & Patelczyk, Maciej)

v3
- Change macro MAX_PTE_PER_SDI from 0x1fe to 0x1feU (De Marchi, Lucas)

Bspec: 60246

Fixes 9c44fd5f6e8a ("drm/xe: Add migrate layer functions for SVM support")
Cc Matthew Brost <[email protected]>
Cc Brian3 Nguyen <[email protected]>
Cc Alex Zuo <[email protected]>
Cc Matthew Auld <[email protected]>
Cc Maciej Patelczyk <[email protected]>
Cc Lucas De Marchi <[email protected]>
(cherry picked from commit c038bdba98c9f6a36378044a9d4385531a194d3e)
---
 drivers/gpu/drm/xe/xe_migrate.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)
drivers/gpu/drm/xe/xe_migrate.c +10 -8
index 5a3e89022c381..752f57fd515e1 100644
@@ -82,7 +82,7 @@ struct xe_migrate {
  * of the instruction.  Subtracting the instruction header (1 dword) and
  * address (2 dwords), that leaves 0x3FD dwords (0x1FE qwords) for PTE values.
  */
-#define MAX_PTE_PER_SDI 0x1FE
+#define MAX_PTE_PER_SDI 0x1FEU
 
 /**
  * xe_tile_migrate_exec_queue() - Get this tile's migrate exec queue.
@@ -1550,15 +1550,17 @@ static u32 pte_update_cmd_size(u64 size)
 	u64 entries = DIV_U64_ROUND_UP(size, XE_PAGE_SIZE);
 
 	XE_WARN_ON(size > MAX_PREEMPTDISABLE_TRANSFER);
+
 	/*
 	 * MI_STORE_DATA_IMM command is used to update page table. Each
-	 * instruction can update maximumly 0x1ff pte entries. To update
-	 * n (n <= 0x1ff) pte entries, we need:
-	 * 1 dword for the MI_STORE_DATA_IMM command header (opcode etc)
-	 * 2 dword for the page table's physical location
-	 * 2*n dword for value of pte to fill (each pte entry is 2 dwords)
+	 * instruction can update maximumly MAX_PTE_PER_SDI pte entries. To
+	 * update n (n <= MAX_PTE_PER_SDI) pte entries, we need:
+	 *
+	 * - 1 dword for the MI_STORE_DATA_IMM command header (opcode etc)
+	 * - 2 dword for the page table's physical location
+	 * - 2*n dword for value of pte to fill (each pte entry is 2 dwords)
 	 */
-	num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, 0x1ff);
+	num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, MAX_PTE_PER_SDI);
 	num_dword += entries * 2;
 
 	return num_dword;
@@ -1574,7 +1576,7 @@ static void build_pt_update_batch_sram(struct xe_migrate *m,
 
 	ptes = DIV_ROUND_UP(size, XE_PAGE_SIZE);
 	while (ptes) {
-		u32 chunk = min(0x1ffU, ptes);
+		u32 chunk = min(MAX_PTE_PER_SDI, ptes);
 
 		bb->cs[bb->len++] = MI_STORE_DATA_IMM | MI_SDI_NUM_QW(chunk);
 		bb->cs[bb->len++] = pt_offset;
Signature · 1 line
2.39.5
author maintainer
#97 lore raw

PATCH 97/178 netfs: Fix hang due to missing case in final DIO read result collection

In reply to Greg Kroah-Hartman

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

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

From: David Howells <[email protected]>

[ Upstream commit da8cf4bd458722d090a788c6e581eeb72695c62f ]

When doing a DIO read, if the subrequests we issue fail and cause the request PAUSE flag to be set to put a pause on subrequest generation, we may complete collection of the subrequests (possibly discarding them) prior to the ALL_QUEUED flags being set.

In such a case, netfs_read_collection() doesn't see ALL_QUEUED being set after netfs_collect_read_results() returns and will just return to the app (the collector can be seen unpausing the generator in the trace log).

The subrequest generator can then set ALL_QUEUED and the app thread reaches netfs_wait_for_request(). This causes netfs_collect_in_app() to be called to see if we're done yet, but there's missing case here.

netfs_collect_in_app() will see that a thread is active and set inactive to false, but won't see any subrequests in the read stream, and so won't set need_collect to true. The function will then just return 0, indicating that the caller should just sleep until further activity (which won't be forthcoming) occurs.

Fix this by making netfs_collect_in_app() check to see if an active thread is complete - i.e. that ALL_QUEUED is set and the subrequests list is empty
- and to skip the sleep return path. The collector will then be called which will clear the request IN_PROGRESS flag, allowing the app to progress.

---
 fs/netfs/misc.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
fs/netfs/misc.c +4 -2
index 43b67a28a8fa0..0a54b12034868 100644
@@ -381,7 +381,7 @@ void netfs_wait_for_in_progress_stream(struct netfs_io_request *rreq,
 static int netfs_collect_in_app(struct netfs_io_request *rreq,
 				bool (*collector)(struct netfs_io_request *rreq))
 {
-	bool need_collect = false, inactive = true;
+	bool need_collect = false, inactive = true, done = true;
 
 	for (int i = 0; i < NR_IO_STREAMS; i++) {
 		struct netfs_io_subrequest *subreq;
@@ -400,9 +400,11 @@ static int netfs_collect_in_app(struct netfs_io_request *rreq,
 			need_collect = true;
 			break;
 		}
+		if (subreq || !test_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags))
+			done = false;
 	}
 
-	if (!need_collect && !inactive)
+	if (!need_collect && !inactive && !done)
 		return 0; /* Sleep */
 
 	__set_current_state(TASK_RUNNING);
Signature · 1 line
2.39.5
author maintainer
#98 lore raw

PATCH 98/178 netfs: Fix looping in wait functions

In reply to Greg Kroah-Hartman

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

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

From: David Howells <[email protected]>

[ Upstream commit 09623e3a14c1c5465124350cd227457c2b0fb017 ]

netfs_wait_for_request() and netfs_wait_for_pause() can loop forever if netfs_collect_in_app() returns 2, indicating that it wants to repeat because the ALL_QUEUED flag isn't yet set and there are no subreqs left that haven't been collected.

The problem is that, unless collection is offloaded (OFFLOAD_COLLECTION), we have to return to the application thread to continue and eventually set ALL_QUEUED after pausing to deal with a retry - but we never get there.

Fix this by inserting checks for the IN_PROGRESS and PAUSE flags as appropriate before cycling round - and add cond_resched() for good measure.

---
 fs/netfs/misc.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)
fs/netfs/misc.c +11 -4
index 0a54b12034868..d8b1a279dbda9 100644
@@ -425,8 +425,8 @@ static int netfs_collect_in_app(struct netfs_io_request *rreq,
 /*
  * Wait for a request to complete, successfully or otherwise.
  */
-static ssize_t netfs_wait_for_request(struct netfs_io_request *rreq,
-				      bool (*collector)(struct netfs_io_request *rreq))
+static ssize_t netfs_wait_for_in_progress(struct netfs_io_request *rreq,
+					  bool (*collector)(struct netfs_io_request *rreq))
 {
 	DEFINE_WAIT(myself);
 	ssize_t ret;
@@ -442,6 +442,9 @@ static ssize_t netfs_wait_for_request(struct netfs_io_request *rreq,
 			case 1:
 				goto all_collected;
 			case 2:
+				if (!netfs_check_rreq_in_progress(rreq))
+					break;
+				cond_resched();
 				continue;
 			}
 		}
@@ -480,12 +483,12 @@ static ssize_t netfs_wait_for_request(struct netfs_io_request *rreq,
 
 ssize_t netfs_wait_for_read(struct netfs_io_request *rreq)
 {
-	return netfs_wait_for_request(rreq, netfs_read_collection);
+	return netfs_wait_for_in_progress(rreq, netfs_read_collection);
 }
 
 ssize_t netfs_wait_for_write(struct netfs_io_request *rreq)
 {
-	return netfs_wait_for_request(rreq, netfs_write_collection);
+	return netfs_wait_for_in_progress(rreq, netfs_write_collection);
 }
 
 /*
@@ -509,6 +512,10 @@ static void netfs_wait_for_pause(struct netfs_io_request *rreq,
 			case 1:
 				goto all_collected;
 			case 2:
+				if (!netfs_check_rreq_in_progress(rreq) ||
+				    !test_bit(NETFS_RREQ_PAUSE, &rreq->flags))
+					break;
+				cond_resched();
 				continue;
 			}
 		}
Signature · 1 line
2.39.5
author maintainer
#72 lore raw

PATCH 72/178 dpaa2-eth: fix xdp_rxq_info leak

In reply to Greg Kroah-Hartman

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

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

From: Fushuai Wang <[email protected]>

[ Upstream commit 2def09ead4ad5907988b655d1e1454003aaf8297 ]

The driver registered xdp_rxq_info structures via xdp_rxq_info_reg() but failed to properly unregister them in error paths and during removal.

---
 .../net/ethernet/freescale/dpaa2/dpaa2-eth.c  | 26 +++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)
drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +24 -2
index 29886a8ba73f3..efd0048acd3b2 100644
@@ -3928,6 +3928,7 @@ static int dpaa2_eth_setup_rx_flow(struct dpaa2_eth_priv *priv,
 					 MEM_TYPE_PAGE_ORDER0, NULL);
 	if (err) {
 		dev_err(dev, "xdp_rxq_info_reg_mem_model failed\n");
+		xdp_rxq_info_unreg(&fq->channel->xdp_rxq);
 		return err;
 	}
 
@@ -4421,17 +4422,25 @@ static int dpaa2_eth_bind_dpni(struct dpaa2_eth_priv *priv)
 			return -EINVAL;
 		}
 		if (err)
-			return err;
+			goto out;
 	}
 
 	err = dpni_get_qdid(priv->mc_io, 0, priv->mc_token,
 			    DPNI_QUEUE_TX, &priv->tx_qdid);
 	if (err) {
 		dev_err(dev, "dpni_get_qdid() failed\n");
-		return err;
+		goto out;
 	}
 
 	return 0;
+
+out:
+	while (i--) {
+		if (priv->fq[i].type == DPAA2_RX_FQ &&
+		    xdp_rxq_info_is_reg(&priv->fq[i].channel->xdp_rxq))
+			xdp_rxq_info_unreg(&priv->fq[i].channel->xdp_rxq);
+	}
+	return err;
 }
 
 /* Allocate rings for storing incoming frame descriptors */
@@ -4814,6 +4823,17 @@ static void dpaa2_eth_del_ch_napi(struct dpaa2_eth_priv *priv)
 	}
 }
 
+static void dpaa2_eth_free_rx_xdp_rxq(struct dpaa2_eth_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->num_fqs; i++) {
+		if (priv->fq[i].type == DPAA2_RX_FQ &&
+		    xdp_rxq_info_is_reg(&priv->fq[i].channel->xdp_rxq))
+			xdp_rxq_info_unreg(&priv->fq[i].channel->xdp_rxq);
+	}
+}
+
 static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev)
 {
 	struct device *dev;
@@ -5017,6 +5037,7 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev)
 	free_percpu(priv->percpu_stats);
 err_alloc_percpu_stats:
 	dpaa2_eth_del_ch_napi(priv);
+	dpaa2_eth_free_rx_xdp_rxq(priv);
 err_bind:
 	dpaa2_eth_free_dpbps(priv);
 err_dpbp_setup:
@@ -5069,6 +5090,7 @@ static void dpaa2_eth_remove(struct fsl_mc_device *ls_dev)
 	free_percpu(priv->percpu_extras);
 
 	dpaa2_eth_del_ch_napi(priv);
+	dpaa2_eth_free_rx_xdp_rxq(priv);
 	dpaa2_eth_free_dpbps(priv);
 	dpaa2_eth_free_dpio(priv);
 	dpaa2_eth_free_dpni(priv);
Signature · 1 line
2.39.5
author maintainer
#99 lore raw

PATCH 99/178 netfs: Fix ref leak on inserted extra subreq in write retry

In reply to Greg Kroah-Hartman

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

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

From: David Howells <[email protected]>

[ Upstream commit 97d8e8e52cb8ab3d7675880a92626d9a4332f7a6 ]

The write-retry algorithm will insert extra subrequests into the list if it can't get sufficient capacity to split the range that needs to be retried into the sequence of subrequests it currently has (for instance, if the cifs credit pool has fewer credits available than it did when the range was originally divided).

However, the allocator furnishes each new subreq with 2 refs and then another is added for resubmission, causing one to be leaked.

Fix this by replacing the ref-getting line with a neutral trace line.

---
 fs/netfs/write_retry.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/netfs/write_retry.c +1 -1
index 9d1d8a8bab726..7158657061e98 100644
@@ -153,7 +153,7 @@ static void netfs_retry_write_stream(struct netfs_io_request *wreq,
 			trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index,
 					     refcount_read(&subreq->ref),
 					     netfs_sreq_trace_new);
-			netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit);
+			trace_netfs_sreq(subreq, netfs_sreq_trace_split);
 
 			list_add(&subreq->rreq_link, &to->rreq_link);
 			to = list_next_entry(to, rreq_link);
Signature · 1 line
2.39.5
author maintainer
#100 lore raw

PATCH 100/178 smb: client: set missing retry flag in smb2_writev_callback()

In reply to Greg Kroah-Hartman

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

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

From: Paulo Alcantara <[email protected]>

[ Upstream commit e67e75edeb88022c04f8e0a173e1ff6dc688f155 ]

Set NETFS_SREQ_NEED_RETRY flag to tell netfslib that the subreq needs to be retried.

---
 fs/smb/client/smb2pdu.c | 1 +
 1 file changed, 1 insertion(+)
fs/smb/client/smb2pdu.c +1 -0
index c3b212175b2bf..2c0cc544dfb31 100644
@@ -4852,6 +4852,7 @@ smb2_writev_callback(struct mid_q_entry *mid)
 		break;
 	case MID_REQUEST_SUBMITTED:
 	case MID_RETRY_NEEDED:
+		__set_bit(NETFS_SREQ_NEED_RETRY, &wdata->subreq.flags);
 		result = -EAGAIN;
 		break;
 	case MID_RESPONSE_MALFORMED:
Signature · 1 line
2.39.5
author maintainer
#101 lore raw

PATCH 101/178 smb: client: set missing retry flag in cifs_readv_callback()

In reply to Greg Kroah-Hartman

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

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

From: Paulo Alcantara <[email protected]>

[ Upstream commit 0e60bae24ad28ab06a485698077d3c626f1e54ab ]

Set NETFS_SREQ_NEED_RETRY flag to tell netfslib that the subreq needs to be retried.

---
 fs/smb/client/cifssmb.c | 1 +
 1 file changed, 1 insertion(+)
fs/smb/client/cifssmb.c +1 -0
index 7216fcec79e8b..f9ccae5de5b88 100644
@@ -1335,6 +1335,7 @@ cifs_readv_callback(struct mid_q_entry *mid)
 		break;
 	case MID_REQUEST_SUBMITTED:
 	case MID_RETRY_NEEDED:
+		__set_bit(NETFS_SREQ_NEED_RETRY, &rdata->subreq.flags);
 		rdata->result = -EAGAIN;
 		if (server->sign && rdata->got_bytes)
 			/* reset bytes number since we can not check a sign */
Signature · 1 line
2.39.5
author maintainer
#73 lore raw

PATCH 73/178 drm/exynos: fimd: Guard display clock control with runtime PM calls

In reply to Greg Kroah-Hartman

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

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

From: Marek Szyprowski <[email protected]>

[ Upstream commit 5d91394f236167ac624b823820faf4aa928b889e ]

Commit c9b1150a68d9 ("drm/atomic-helper: Re-order bridge chain pre-enable and post-disable") changed the call sequence to the CRTC enable/disable and bridge pre_enable/post_disable methods, so those bridge methods are now called when CRTC is not yet enabled.

This causes a lockup observed on Samsung Peach-Pit/Pi Chromebooks. The source of this lockup is a call to fimd_dp_clock_enable() function, when FIMD device is not yet runtime resumed. It worked before the mentioned commit only because the CRTC implemented by the FIMD driver was always enabled what guaranteed the FIMD device to be runtime resumed.

This patch adds runtime PM guards to the fimd_dp_clock_enable() function to enable its proper operation also when the CRTC implemented by FIMD is not yet enabled.

Fixes 196e059a8a6a ("drm/exynos: convert clock_enable crtc callback to pipeline clock")
---
 drivers/gpu/drm/exynos/exynos_drm_fimd.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
drivers/gpu/drm/exynos/exynos_drm_fimd.c +12 -0
index c394cc702d7d4..205c238cc73a6 100644
@@ -187,6 +187,7 @@ struct fimd_context {
 	u32				i80ifcon;
 	bool				i80_if;
 	bool				suspended;
+	bool				dp_clk_enabled;
 	wait_queue_head_t		wait_vsync_queue;
 	atomic_t			wait_vsync_event;
 	atomic_t			win_updated;
@@ -1047,7 +1048,18 @@ static void fimd_dp_clock_enable(struct exynos_drm_clk *clk, bool enable)
 	struct fimd_context *ctx = container_of(clk, struct fimd_context,
 						dp_clk);
 	u32 val = enable ? DP_MIE_CLK_DP_ENABLE : DP_MIE_CLK_DISABLE;
+
+	if (enable == ctx->dp_clk_enabled)
+		return;
+
+	if (enable)
+		pm_runtime_resume_and_get(ctx->dev);
+
+	ctx->dp_clk_enabled = enable;
 	writel(val, ctx->regs + DP_MIE_CLKCON);
+
+	if (!enable)
+		pm_runtime_put(ctx->dev);
 }
 
 static const struct exynos_drm_crtc_ops fimd_crtc_ops = {
Signature · 1 line
2.39.5
author maintainer
#74 lore raw

PATCH 74/178 spi: spi-qpic-snand: reallocate BAM transactions

In reply to Greg Kroah-Hartman

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

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

From: Gabor Juhos <[email protected]>

[ Upstream commit d85d0380292a7e618915069c3579ae23c7c80339 ]

Using the mtd_nandbiterrs module for testing the driver occasionally results in weird things like below.

1. swiotlb mapping fails with the following message:

  [   85.926216] qcom_snand 79b0000.spi: swiotlb buffer is full (sz: 4294967294 bytes), total 512 (slots), used 0 (slots)
  [   85.932937] qcom_snand 79b0000.spi: failure in mapping desc
  [   87.999314] qcom_snand 79b0000.spi: failure to write raw page
  [   87.999352] mtd_nandbiterrs: error: write_oob failed (-110)

Rebooting the board after this causes a panic due to a NULL pointer dereference.

2. If the swiotlb mapping does not fail, rebooting the board may result in a different panic due to a bad spinlock magic:

  [  256.104459] BUG: spinlock bad magic on CPU#3, procd/2241
  [  256.104488] Unable to handle kernel paging request at virtual address ffffffff0000049b
  ...

Investigating the issue revealed that these symptoms are results of memory corruption which is caused by out of bounds access within the driver.

The driver uses a dynamically allocated structure for BAM transactions, which structure must have enough space for all possible variations of different flash operations initiated by the driver. The required space heavily depends on the actual number of 'codewords' which is calculated from the pagesize of the actual NAND chip.

Although the qcom_nandc_alloc() function allocates memory for the BAM transactions during probe, but since the actual number of 'codewords' is not yet know the allocation is done for one 'codeword' only.

Because of this, whenever the driver does a flash operation, and the number of the required transactions exceeds the size of the allocated arrays the driver accesses memory out of the allocated range.

To avoid this, change the code to free the initially allocated BAM transactions memory, and allocate a new one once the actual number of 'codewords' required for a given NAND chip is known.

---
 drivers/spi/spi-qpic-snand.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/spi/spi-qpic-snand.c +16 -0
index d3a4e091dca4e..80856d2fa35c6 100644
@@ -316,6 +316,22 @@ static int qcom_spi_ecc_init_ctx_pipelined(struct nand_device *nand)
 
 	mtd_set_ooblayout(mtd, &qcom_spi_ooblayout);
 
+	/*
+	 * Free the temporary BAM transaction allocated initially by
+	 * qcom_nandc_alloc(), and allocate a new one based on the
+	 * updated max_cwperpage value.
+	 */
+	qcom_free_bam_transaction(snandc);
+
+	snandc->max_cwperpage = cwperpage;
+
+	snandc->bam_txn = qcom_alloc_bam_transaction(snandc);
+	if (!snandc->bam_txn) {
+		dev_err(snandc->dev, "failed to allocate BAM transaction\n");
+		ret = -ENOMEM;
+		goto err_free_ecc_cfg;
+	}
+
 	ecc_cfg->cfg0 = FIELD_PREP(CW_PER_PAGE_MASK, (cwperpage - 1)) |
 			FIELD_PREP(UD_SIZE_BYTES_MASK, ecc_cfg->cw_data) |
 			FIELD_PREP(DISABLE_STATUS_AFTER_WRITE, 1) |
Signature · 1 line
2.39.5
author maintainer
#75 lore raw

PATCH 75/178 spi: spi-fsl-dspi: Clear completion counter before initiating transfer

In reply to Greg Kroah-Hartman

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

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

From: James Clark <[email protected]>

[ Upstream commit fa60c094c19b97e103d653f528f8d9c178b6a5f5 ]

In target mode, extra interrupts can be received between the end of a transfer and halting the module if the host continues sending more data. If the interrupt from this occurs after the reinit_completion() then the completion counter is left at a non-zero value. The next unrelated transfer initiated by userspace will then complete immediately without waiting for the interrupt or writing to the RX buffer.

Fix it by resetting the counter before the transfer so that lingering values are cleared. This is done after clearing the FIFOs and the status register but before the transfer is initiated, so no interrupts should be received at this point resulting in other race conditions.

Fixes 4f5ee75ea171 ("spi: spi-fsl-dspi: Replace interruptible wait queue with a simple completion")
---
 drivers/spi/spi-fsl-dspi.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
drivers/spi/spi-fsl-dspi.c +10 -1
index 863781ba6c160..0dcd491140950 100644
@@ -983,11 +983,20 @@ static int dspi_transfer_one_message(struct spi_controller *ctlr,
 		if (dspi->devtype_data->trans_mode == DSPI_DMA_MODE) {
 			status = dspi_dma_xfer(dspi);
 		} else {
+			/*
+			 * Reinitialize the completion before transferring data
+			 * to avoid the case where it might remain in the done
+			 * state due to a spurious interrupt from a previous
+			 * transfer. This could falsely signal that the current
+			 * transfer has completed.
+			 */
+			if (dspi->irq)
+				reinit_completion(&dspi->xfer_done);
+
 			dspi_fifo_write(dspi);
 
 			if (dspi->irq) {
 				wait_for_completion(&dspi->xfer_done);
-				reinit_completion(&dspi->xfer_done);
 			} else {
 				do {
 					status = dspi_poll(dspi);
Signature · 1 line
2.39.5
author maintainer
#76 lore raw

PATCH 76/178 drm/i915/selftests: Change mock_request() to return error pointers

In reply to Greg Kroah-Hartman

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

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

From: Dan Carpenter <[email protected]>

[ Upstream commit caa7c7a76b78ce41d347003f84975125383e6b59 ]

There was an error pointer vs NULL bug in __igt_breadcrumbs_smoketest(). The __mock_request_alloc() function implements the smoketest->request_alloc() function pointer. It was supposed to return error pointers, but it propogates the NULL return from mock_request() so in the event of a failure, it would lead to a NULL pointer dereference.

To fix this, change the mock_request() function to return error pointers and update all the callers to expect that.

Fixes 52c0fdb25c7c ("drm/i915: Replace global breadcrumbs with per-context interrupt tracking")
(cherry picked from commit 778fa8ad5f0f23397d045c7ebca048ce8def1c43)
---
 drivers/gpu/drm/i915/selftests/i915_request.c | 20 +++++++++----------
 drivers/gpu/drm/i915/selftests/mock_request.c |  2 +-
 2 files changed, 11 insertions(+), 11 deletions(-)
drivers/gpu/drm/i915/selftests/i915_request.c +10 -10
index 88870844b5bd9..2fb7a9e7efec6 100644
@@ -73,8 +73,8 @@ static int igt_add_request(void *arg)
 	/* Basic preliminary test to create a request and let it loose! */
 
 	request = mock_request(rcs0(i915)->kernel_context, HZ / 10);
-	if (!request)
-		return -ENOMEM;
+	if (IS_ERR(request))
+		return PTR_ERR(request);
 
 	i915_request_add(request);
 
@@ -91,8 +91,8 @@ static int igt_wait_request(void *arg)
 	/* Submit a request, then wait upon it */
 
 	request = mock_request(rcs0(i915)->kernel_context, T);
-	if (!request)
-		return -ENOMEM;
+	if (IS_ERR(request))
+		return PTR_ERR(request);
 
 	i915_request_get(request);
 
@@ -160,8 +160,8 @@ static int igt_fence_wait(void *arg)
 	/* Submit a request, treat it as a fence and wait upon it */
 
 	request = mock_request(rcs0(i915)->kernel_context, T);
-	if (!request)
-		return -ENOMEM;
+	if (IS_ERR(request))
+		return PTR_ERR(request);
 
 	if (dma_fence_wait_timeout(&request->fence, false, T) != -ETIME) {
 		pr_err("fence wait success before submit (expected timeout)!\n");
@@ -219,8 +219,8 @@ static int igt_request_rewind(void *arg)
 	GEM_BUG_ON(IS_ERR(ce));
 	request = mock_request(ce, 2 * HZ);
 	intel_context_put(ce);
-	if (!request) {
-		err = -ENOMEM;
+	if (IS_ERR(request)) {
+		err = PTR_ERR(request);
 		goto err_context_0;
 	}
 
@@ -237,8 +237,8 @@ static int igt_request_rewind(void *arg)
 	GEM_BUG_ON(IS_ERR(ce));
 	vip = mock_request(ce, 0);
 	intel_context_put(ce);
-	if (!vip) {
-		err = -ENOMEM;
+	if (IS_ERR(vip)) {
+		err = PTR_ERR(vip);
 		goto err_context_1;
 	}
 
drivers/gpu/drm/i915/selftests/mock_request.c +1 -1
index 09f747228dff5..1b0cf073e9643 100644
@@ -35,7 +35,7 @@ mock_request(struct intel_context *ce, unsigned long delay)
 	/* NB the i915->requests slab cache is enlarged to fit mock_request */
 	request = intel_context_create_request(ce);
 	if (IS_ERR(request))
-		return NULL;
+		return request;
 
 	request->mock.delay = delay;
 	return request;
Signature · 1 line
2.39.5
author maintainer
#77 lore raw

PATCH 77/178 nvme: Fix incorrect cdw15 value in passthru error logging

In reply to Greg Kroah-Hartman

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

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

From: Alok Tiwari <[email protected]>

[ Upstream commit 2e96d2d8c2a7a6c2cef45593c028d9c5ef180316 ]

Fix an error in nvme_log_err_passthru() where cdw14 was incorrectly printed twice instead of cdw15. This fix ensures accurate logging of the full passthrough command payload.

Fixes 9f079dda1433 ("nvme: allow passthru cmd error logging")
---
 drivers/nvme/host/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/nvme/host/core.c +1 -1
index d253b82901110..ae584c97f5284 100644
@@ -382,7 +382,7 @@ static void nvme_log_err_passthru(struct request *req)
 		nr->cmd->common.cdw12,
 		nr->cmd->common.cdw13,
 		nr->cmd->common.cdw14,
-		nr->cmd->common.cdw14);
+		nr->cmd->common.cdw15);
 }
 
 enum nvme_disposition {
Signature · 1 line
2.39.5
author maintainer
#78 lore raw

PATCH 78/178 nvmet: fix memory leak of bio integrity

In reply to Greg Kroah-Hartman

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

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

From: Dmitry Bogdanov <[email protected]>

[ Upstream commit 190f4c2c863af7cc5bb354b70e0805f06419c038 ]

If nvmet receives commands with metadata there is a continuous memory leak of kmalloc-128 slab or more precisely bio->bi_integrity.

Since commit bf4c89fc8797 ("block: don't call bio_uninit from bio_endio") each user of bio_init has to use bio_uninit as well. Otherwise the bio integrity is not getting free. Nvmet uses bio_init for inline bios.

Uninit the inline bio to complete deallocation of integrity in bio.

Fixes bf4c89fc8797 ("block: don't call bio_uninit from bio_endio")
---
 drivers/nvme/target/nvmet.h | 2 ++
 1 file changed, 2 insertions(+)
drivers/nvme/target/nvmet.h +2 -0
index b6db8b74dc4ad..85e50e2981742 100644
@@ -857,6 +857,8 @@ static inline void nvmet_req_bio_put(struct nvmet_req *req, struct bio *bio)
 {
 	if (bio != &req->b.inline_bio)
 		bio_put(bio);
+	else
+		bio_uninit(bio);
 }
 
 #ifdef CONFIG_NVME_TARGET_TCP_TLS
Signature · 1 line
2.39.5
author maintainer
#69 lore raw

PATCH 69/178 btrfs: propagate last_unlink_trans earlier when doing a rmdir

In reply to Greg Kroah-Hartman

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

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

From: Filipe Manana <[email protected]>

[ Upstream commit c466e33e729a0ee017d10d919cba18f503853c60 ]

In case the removed directory had a snapshot that was deleted, we are propagating its inode's last_unlink_trans to the parent directory after we removed the entry from the parent directory. This leaves a small race window where someone can log the parent directory after we removed the entry and before we updated last_unlink_trans, and as a result if we ever try to replay such a log tree, we will fail since we will attempt to remove a snapshot during log replay, which is currently not possible and results in the log replay (and mount) to fail. This is the type of failure described in commit 1ec9a1ae1e30 ("Btrfs: fix unreplayable log after snapshot delete + parent dir fsync").

So fix this by propagating the last_unlink_trans to the parent directory before we remove the entry from it.

Fixes 44f714dae50a ("Btrfs: improve performance on fsync against new inode after rename/unlink")
---
 fs/btrfs/inode.c | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)
fs/btrfs/inode.c +18 -18
index 391172b443e50..24db00a9319b3 100644
@@ -4724,7 +4724,6 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
 	int ret = 0;
 	struct btrfs_trans_handle *trans;
-	u64 last_unlink_trans;
 	struct fscrypt_name fname;
 
 	if (inode->i_size > BTRFS_EMPTY_DIR_SIZE)
@@ -4750,6 +4749,23 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 		goto out_notrans;
 	}
 
+	/*
+	 * Propagate the last_unlink_trans value of the deleted dir to its
+	 * parent directory. This is to prevent an unrecoverable log tree in the
+	 * case we do something like this:
+	 * 1) create dir foo
+	 * 2) create snapshot under dir foo
+	 * 3) delete the snapshot
+	 * 4) rmdir foo
+	 * 5) mkdir foo
+	 * 6) fsync foo or some file inside foo
+	 *
+	 * This is because we can't unlink other roots when replaying the dir
+	 * deletes for directory foo.
+	 */
+	if (BTRFS_I(inode)->last_unlink_trans >= trans->transid)
+		BTRFS_I(dir)->last_unlink_trans = BTRFS_I(inode)->last_unlink_trans;
+
 	if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
 		ret = btrfs_unlink_subvol(trans, BTRFS_I(dir), dentry);
 		goto out;
@@ -4759,27 +4775,11 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 	if (ret)
 		goto out;
 
-	last_unlink_trans = BTRFS_I(inode)->last_unlink_trans;
-
 	/* now the directory is empty */
 	ret = btrfs_unlink_inode(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)),
 				 &fname.disk_name);
-	if (!ret) {
+	if (!ret)
 		btrfs_i_size_write(BTRFS_I(inode), 0);
-		/*
-		 * Propagate the last_unlink_trans value of the deleted dir to
-		 * its parent directory. This is to prevent an unrecoverable
-		 * log tree in the case we do something like this:
-		 * 1) create dir foo
-		 * 2) create snapshot under dir foo
-		 * 3) delete the snapshot
-		 * 4) rmdir foo
-		 * 5) mkdir foo
-		 * 6) fsync foo or some file inside foo
-		 */
-		if (last_unlink_trans >= trans->transid)
-			BTRFS_I(dir)->last_unlink_trans = last_unlink_trans;
-	}
 out:
 	btrfs_end_transaction(trans);
 out_notrans:
Signature · 1 line
2.39.5
author maintainer
#103 lore raw

PATCH 103/178 netfs: Fix i_size updating

In reply to Greg Kroah-Hartman

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

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

From: David Howells <[email protected]>

[ Upstream commit 2e0658940d90a3dc130bb3b7f75bae9f4100e01f ]

Fix the updating of i_size, particularly in regard to the completion of DIO writes and especially async DIO writes by using a lock.

The bug is triggered occasionally by the generic/207 xfstest as it chucks a bunch of AIO DIO writes at the filesystem and then checks that fstat() returns a reasonable st_size as each completes.

The problem is that netfs is trying to do "if new_size > inode->i_size, update inode->i_size" sort of thing but without a lock around it.

This can be seen with cifs, but shouldn't be seen with kafs because kafs serialises modification ops on the client whereas cifs sends the requests to the server as they're generated and lets the server order them.

---
 fs/netfs/buffered_write.c | 2 ++
 fs/netfs/direct_write.c   | 8 ++++++--
 2 files changed, 8 insertions(+), 2 deletions(-)
fs/netfs/buffered_write.c +2 -0
index dbb544e183d13..9f22ff890a8cd 100644
@@ -64,6 +64,7 @@ static void netfs_update_i_size(struct netfs_inode *ctx, struct inode *inode,
 		return;
 	}
 
+	spin_lock(&inode->i_lock);
 	i_size_write(inode, pos);
 #if IS_ENABLED(CONFIG_FSCACHE)
 	fscache_update_cookie(ctx->cache, NULL, &pos);
@@ -77,6 +78,7 @@ static void netfs_update_i_size(struct netfs_inode *ctx, struct inode *inode,
 					DIV_ROUND_UP(pos, SECTOR_SIZE),
 					inode->i_blocks + add);
 	}
+	spin_unlock(&inode->i_lock);
 }
 
 /**
fs/netfs/direct_write.c +6 -2
index fa9a5bf3c6d51..3efa5894b2c07 100644
@@ -14,13 +14,17 @@ static void netfs_cleanup_dio_write(struct netfs_io_request *wreq)
 	struct inode *inode = wreq->inode;
 	unsigned long long end = wreq->start + wreq->transferred;
 
-	if (!wreq->error &&
-	    i_size_read(inode) < end) {
+	if (wreq->error || end <= i_size_read(inode))
+		return;
+
+	spin_lock(&inode->i_lock);
+	if (end > i_size_read(inode)) {
 		if (wreq->netfs_ops->update_i_size)
 			wreq->netfs_ops->update_i_size(inode, end);
 		else
 			i_size_write(inode, end);
 	}
+	spin_unlock(&inode->i_lock);
 }
 
 /*
Signature · 1 line
2.39.5
author maintainer
#112 lore raw

PATCH 112/178 rose: fix dangling neighbour pointers in rose_rt_device_down()

In reply to Greg Kroah-Hartman

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

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

From: Kohei Enju <[email protected]>

[ Upstream commit 34a500caf48c47d5171f4aa1f237da39b07c6157 ]

There are two bugs in rose_rt_device_down() that can cause use-after-free:

1. The loop bound t->count is modified within the loop, which can cause the loop to terminate early and miss some entries.

2. When removing an entry from the neighbour array, the subsequent entries are moved up to fill the gap, but the loop index i is still incremented, causing the next entry to be skipped.

For example, if a node has three neighbours (A, A, B) with count=3 and A is being removed, the second A is not checked.

i=0: (A, A, B) -> (A, B) with count=2
^ checked
i=1: (A, B) -> (A, B) with count=2
^ checked (B, not A!)
i=2: (doesn't occur because i < count is false)

This leaves the second A in the array with count=2, but the rose_neigh structure has been freed. Code that accesses these entries assumes that the first count entries are valid pointers, causing a use-after-free when it accesses the dangling pointer.

Fix both issues by iterating over the array in reverse order with a fixed loop bound. This ensures that all entries are examined and that the removal of an entry doesn't affect subsequent iterations.

---
 net/rose/rose_route.c | 15 ++++-----------
 1 file changed, 4 insertions(+), 11 deletions(-)
net/rose/rose_route.c +4 -11
index 2dd6bd3a3011f..b72bf8a08d489 100644
@@ -497,22 +497,15 @@ void rose_rt_device_down(struct net_device *dev)
 			t         = rose_node;
 			rose_node = rose_node->next;
 
-			for (i = 0; i < t->count; i++) {
+			for (i = t->count - 1; i >= 0; i--) {
 				if (t->neighbour[i] != s)
 					continue;
 
 				t->count--;
 
-				switch (i) {
-				case 0:
-					t->neighbour[0] = t->neighbour[1];
-					fallthrough;
-				case 1:
-					t->neighbour[1] = t->neighbour[2];
-					break;
-				case 2:
-					break;
-				}
+				memmove(&t->neighbour[i], &t->neighbour[i + 1],
+					sizeof(t->neighbour[0]) *
+						(t->count - i));
 			}
 
 			if (t->count <= 0)
Signature · 1 line
2.39.5
author maintainer
#113 lore raw

PATCH 113/178 nui: Fix dma_mapping_error() check

In reply to Greg Kroah-Hartman

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

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

From: Thomas Fourier <[email protected]>

[ Upstream commit 561aa0e22b70a5e7246b73d62a824b3aef3fc375 ]

dma_map_XXX() functions return values DMA_MAPPING_ERROR as error values which is often ~0. The error value should be tested with dma_mapping_error().

This patch creates a new function in niu_ops to test if the mapping failed. The test is fixed in niu_rbr_add_page(), added in niu_start_xmit() and the successfully mapped pages are unmaped upon error.

Fixes ec2deec1f352 ("niu: Fix to check for dma mapping errors.")
---
 drivers/net/ethernet/sun/niu.c | 31 ++++++++++++++++++++++++++++++-
 drivers/net/ethernet/sun/niu.h |  4 ++++
 2 files changed, 34 insertions(+), 1 deletion(-)
drivers/net/ethernet/sun/niu.c +30 -1
index 379b6e90121d9..02b8a9e5c9a90 100644
@@ -3336,7 +3336,7 @@ static int niu_rbr_add_page(struct niu *np, struct rx_ring_info *rp,
 
 	addr = np->ops->map_page(np->device, page, 0,
 				 PAGE_SIZE, DMA_FROM_DEVICE);
-	if (!addr) {
+	if (np->ops->mapping_error(np->device, addr)) {
 		__free_page(page);
 		return -ENOMEM;
 	}
@@ -6676,6 +6676,8 @@ static netdev_tx_t niu_start_xmit(struct sk_buff *skb,
 	len = skb_headlen(skb);
 	mapping = np->ops->map_single(np->device, skb->data,
 				      len, DMA_TO_DEVICE);
+	if (np->ops->mapping_error(np->device, mapping))
+		goto out_drop;
 
 	prod = rp->prod;
 
@@ -6717,6 +6719,8 @@ static netdev_tx_t niu_start_xmit(struct sk_buff *skb,
 		mapping = np->ops->map_page(np->device, skb_frag_page(frag),
 					    skb_frag_off(frag), len,
 					    DMA_TO_DEVICE);
+		if (np->ops->mapping_error(np->device, mapping))
+			goto out_unmap;
 
 		rp->tx_buffs[prod].skb = NULL;
 		rp->tx_buffs[prod].mapping = mapping;
@@ -6741,6 +6745,19 @@ static netdev_tx_t niu_start_xmit(struct sk_buff *skb,
 out:
 	return NETDEV_TX_OK;
 
+out_unmap:
+	while (i--) {
+		const skb_frag_t *frag;
+
+		prod = PREVIOUS_TX(rp, prod);
+		frag = &skb_shinfo(skb)->frags[i];
+		np->ops->unmap_page(np->device, rp->tx_buffs[prod].mapping,
+				    skb_frag_size(frag), DMA_TO_DEVICE);
+	}
+
+	np->ops->unmap_single(np->device, rp->tx_buffs[rp->prod].mapping,
+			      skb_headlen(skb), DMA_TO_DEVICE);
+
 out_drop:
 	rp->tx_errors++;
 	kfree_skb(skb);
@@ -9644,6 +9661,11 @@ static void niu_pci_unmap_single(struct device *dev, u64 dma_address,
 	dma_unmap_single(dev, dma_address, size, direction);
 }
 
+static int niu_pci_mapping_error(struct device *dev, u64 addr)
+{
+	return dma_mapping_error(dev, addr);
+}
+
 static const struct niu_ops niu_pci_ops = {
 	.alloc_coherent	= niu_pci_alloc_coherent,
 	.free_coherent	= niu_pci_free_coherent,
@@ -9651,6 +9673,7 @@ static const struct niu_ops niu_pci_ops = {
 	.unmap_page	= niu_pci_unmap_page,
 	.map_single	= niu_pci_map_single,
 	.unmap_single	= niu_pci_unmap_single,
+	.mapping_error	= niu_pci_mapping_error,
 };
 
 static void niu_driver_version(void)
@@ -10019,6 +10042,11 @@ static void niu_phys_unmap_single(struct device *dev, u64 dma_address,
 	/* Nothing to do.  */
 }
 
+static int niu_phys_mapping_error(struct device *dev, u64 dma_address)
+{
+	return false;
+}
+
 static const struct niu_ops niu_phys_ops = {
 	.alloc_coherent	= niu_phys_alloc_coherent,
 	.free_coherent	= niu_phys_free_coherent,
@@ -10026,6 +10054,7 @@ static const struct niu_ops niu_phys_ops = {
 	.unmap_page	= niu_phys_unmap_page,
 	.map_single	= niu_phys_map_single,
 	.unmap_single	= niu_phys_unmap_single,
+	.mapping_error	= niu_phys_mapping_error,
 };
 
 static int niu_of_probe(struct platform_device *op)
drivers/net/ethernet/sun/niu.h +4 -0
index 04c215f91fc08..0b169c08b0f2d 100644
@@ -2879,6 +2879,9 @@ struct tx_ring_info {
 #define NEXT_TX(tp, index) \
 	(((index) + 1) < (tp)->pending ? ((index) + 1) : 0)
 
+#define PREVIOUS_TX(tp, index) \
+	(((index) - 1) >= 0 ? ((index) - 1) : (((tp)->pending) - 1))
+
 static inline u32 niu_tx_avail(struct tx_ring_info *tp)
 {
 	return (tp->pending -
@@ -3140,6 +3143,7 @@ struct niu_ops {
 			  enum dma_data_direction direction);
 	void (*unmap_single)(struct device *dev, u64 dma_address,
 			     size_t size, enum dma_data_direction direction);
+	int (*mapping_error)(struct device *dev, u64 dma_address);
 };
 
 struct niu_link_config {
Signature · 1 line
2.39.5
author maintainer
#114 lore raw

PATCH 114/178 net/sched: Always pass notifications when child class becomes empty

In reply to Greg Kroah-Hartman

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

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

From: Lion Ackermann <[email protected]>

[ Upstream commit 103406b38c600fec1fe375a77b27d87e314aea09 ]

Certain classful qdiscs may invoke their classes' dequeue handler on an enqueue operation. This may unexpectedly empty the child qdisc and thus make an in-flight class passive via qlen_notify(). Most qdiscs do not expect such behaviour at this point in time and may re-activate the class eventually anyways which will lead to a use-after-free.

The referenced fix commit attempted to fix this behavior for the HFSC case by moving the backlog accounting around, though this turned out to be incomplete since the parent's parent may run into the issue too. The following reproducer demonstrates this use-after-free:

tc qdisc add dev lo root handle 1: drr
tc filter add dev lo parent 1: basic classid 1:1
tc class add dev lo parent 1: classid 1:1 drr
tc qdisc add dev lo parent 1:1 handle 2: hfsc def 1 tc class add dev lo parent 2: classid 2:1 hfsc rt m1 8 d 1 m2 0 tc qdisc add dev lo parent 2:1 handle 3: netem
tc qdisc add dev lo parent 3:1 handle 4: blackhole

echo 1 | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888 tc class delete dev lo classid 1:1
echo 1 | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888

Since backlog accounting issues leading to a use-after-frees on stale class pointers is a recurring pattern at this point, this patch takes a different approach. Instead of trying to fix the accounting, the patch ensures that qdisc_tree_reduce_backlog always calls qlen_notify when the child qdisc is empty. This solves the problem because deletion of qdiscs always involves a call to qdisc_reset() and / or qdisc_purge_queue() which ultimately resets its qlen to 0 thus causing the following qdisc_tree_reduce_backlog() to report to the parent. Note that this may call qlen_notify on passive classes multiple times. This is not a problem after the recent patch series that made all the classful qdiscs qlen_notify() handlers idempotent.

---
 net/sched/sch_api.c | 19 +++++--------------
 1 file changed, 5 insertions(+), 14 deletions(-)
net/sched/sch_api.c +5 -14
index f74a097f54ae7..d58921ffcf35e 100644
@@ -779,15 +779,12 @@ static u32 qdisc_alloc_handle(struct net_device *dev)
 
 void qdisc_tree_reduce_backlog(struct Qdisc *sch, int n, int len)
 {
-	bool qdisc_is_offloaded = sch->flags & TCQ_F_OFFLOADED;
 	const struct Qdisc_class_ops *cops;
 	unsigned long cl;
 	u32 parentid;
 	bool notify;
 	int drops;
 
-	if (n == 0 && len == 0)
-		return;
 	drops = max_t(int, n, 0);
 	rcu_read_lock();
 	while ((parentid = sch->parent)) {
@@ -796,17 +793,8 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, int n, int len)
 
 		if (sch->flags & TCQ_F_NOPARENT)
 			break;
-		/* Notify parent qdisc only if child qdisc becomes empty.
-		 *
-		 * If child was empty even before update then backlog
-		 * counter is screwed and we skip notification because
-		 * parent class is already passive.
-		 *
-		 * If the original child was offloaded then it is allowed
-		 * to be seem as empty, so the parent is notified anyway.
-		 */
-		notify = !sch->q.qlen && !WARN_ON_ONCE(!n &&
-						       !qdisc_is_offloaded);
+		/* Notify parent qdisc only if child qdisc becomes empty. */
+		notify = !sch->q.qlen;
 		/* TODO: perform the search on a per txq basis */
 		sch = qdisc_lookup_rcu(qdisc_dev(sch), TC_H_MAJ(parentid));
 		if (sch == NULL) {
@@ -815,6 +803,9 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, int n, int len)
 		}
 		cops = sch->ops->cl_ops;
 		if (notify && cops->qlen_notify) {
+			/* Note that qlen_notify must be idempotent as it may get called
+			 * multiple times.
+			 */
 			cl = cops->find(sch, parentid);
 			cops->qlen_notify(sch, cl);
 		}
Signature · 1 line
2.39.5
author maintainer
#115 lore raw

PATCH 115/178 amd-xgbe: do not double read link status

In reply to Greg Kroah-Hartman

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

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

From: Raju Rangoju <[email protected]>

[ Upstream commit 16ceda2ef683a50cd0783006c0504e1931cd8879 ]

The link status is latched low so that momentary link drops can be detected. Always double-reading the status defeats this design feature. Only double read if link was already down

This prevents unnecessary duplicate readings of the link status.

---
 drivers/net/ethernet/amd/xgbe/xgbe-mdio.c   |  4 ++++
 drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c | 24 +++++++++++++--------
 2 files changed, 19 insertions(+), 9 deletions(-)
drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +4 -0
index 3316c719f9f8c..ed76a8df6ec6e 100644
@@ -1413,6 +1413,10 @@ static void xgbe_phy_status(struct xgbe_prv_data *pdata)
 
 	pdata->phy.link = pdata->phy_if.phy_impl.link_status(pdata,
 							     &an_restart);
+	/* bail out if the link status register read fails */
+	if (pdata->phy.link < 0)
+		return;
+
 	if (an_restart) {
 		xgbe_phy_config_aneg(pdata);
 		goto adjust_link;
drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c +15 -9
index 268399dfcf22f..32e633d113484 100644
@@ -2855,8 +2855,7 @@ static bool xgbe_phy_valid_speed(struct xgbe_prv_data *pdata, int speed)
 static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart)
 {
 	struct xgbe_phy_data *phy_data = pdata->phy_data;
-	unsigned int reg;
-	int ret;
+	int reg, ret;
 
 	*an_restart = 0;
 
@@ -2890,11 +2889,20 @@ static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart)
 			return 0;
 	}
 
-	/* Link status is latched low, so read once to clear
-	 * and then read again to get current state
-	 */
-	reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
 	reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
+	if (reg < 0)
+		return reg;
+
+	/* Link status is latched low so that momentary link drops
+	 * can be detected. If link was already down read again
+	 * to get the latest state.
+	 */
+
+	if (!pdata->phy.link && !(reg & MDIO_STAT1_LSTATUS)) {
+		reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
+		if (reg < 0)
+			return reg;
+	}
 
 	if (pdata->en_rx_adap) {
 		/* if the link is available and adaptation is done,
@@ -2913,9 +2921,7 @@ static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart)
 			xgbe_phy_set_mode(pdata, phy_data->cur_mode);
 		}
 
-		/* check again for the link and adaptation status */
-		reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1);
-		if ((reg & MDIO_STAT1_LSTATUS) && pdata->rx_adapt_done)
+		if (pdata->rx_adapt_done)
 			return 1;
 	} else if (reg & MDIO_STAT1_LSTATUS)
 		return 1;
Signature · 1 line
2.39.5
author maintainer
#116 lore raw

PATCH 116/178 net: ipv4: fix stat increase when udp early demux drops the packet

In reply to Greg Kroah-Hartman

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

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

From: Antoine Tenart <[email protected]>

[ Upstream commit c2a2ff6b4db55647575260bf2227b0e09d46addb ]

udp_v4_early_demux now returns drop reasons as it either returns 0 or ip_mc_validate_source, which returns itself a drop reason. However its use was not converted in ip_rcv_finish_core and the drop reason is ignored, leading to potentially skipping increasing LINUX_MIB_IPRPFILTER if the drop reason is SKB_DROP_REASON_IP_RPFILTER.

This is a fix and we're not converting udp_v4_early_demux to explicitly return a drop reason to ease backports; this can be done as a follow-up.

Fixes d46f827016d8 ("net: ip: make ip_mc_validate_source() return drop reason")
Cc Menglong Dong <[email protected]>
---
 net/ipv4/ip_input.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
net/ipv4/ip_input.c +4 -3
index 30a5e9460d006..5a49eb99e5c48 100644
@@ -319,8 +319,8 @@ static int ip_rcv_finish_core(struct net *net,
 			      const struct sk_buff *hint)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	int err, drop_reason;
 	struct rtable *rt;
+	int drop_reason;
 
 	if (ip_can_use_hint(skb, iph, hint)) {
 		drop_reason = ip_route_use_hint(skb, iph->daddr, iph->saddr,
@@ -345,9 +345,10 @@ static int ip_rcv_finish_core(struct net *net,
 			break;
 		case IPPROTO_UDP:
 			if (READ_ONCE(net->ipv4.sysctl_udp_early_demux)) {
-				err = udp_v4_early_demux(skb);
-				if (unlikely(err))
+				drop_reason = udp_v4_early_demux(skb);
+				if (unlikely(drop_reason))
 					goto drop_error;
+				drop_reason = SKB_DROP_REASON_NOT_SPECIFIED;
 
 				/* must reload iph, skb->head might have changed */
 				iph = ip_hdr(skb);
Signature · 1 line
2.39.5
author maintainer
#117 lore raw

PATCH 117/178 smb: client: fix race condition in negotiate timeout by using more precise timing

In reply to Greg Kroah-Hartman

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

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

From: Wang Zhaolong <[email protected]>

[ Upstream commit 266b5d02e14f3a0e07414e11f239397de0577a1d ]

When the SMB server reboots and the client immediately accesses the mount point, a race condition can occur that causes operations to fail with "Host is down" error.

Reproduction steps:
# Mount SMB share
mount -t cifs //192.168.245.109/TEST /mnt/ -o xxxx ls /mnt

# Reboot server
ssh [email protected] reboot
ssh [email protected] /path/to/cifs_server_setup.sh ssh [email protected] systemctl stop firewalld

# Immediate access fails
ls /mnt
ls: cannot access '/mnt': Host is down

# But works if there is a delay

The issue is caused by a race condition between negotiate and reconnect. The 20-second negotiate timeout mechanism can interfere with the normal recovery process when both are triggered simultaneously.

ls cifsd
---------------------------------------------------
cifs_getattr
cifs_revalidate_dentry
cifs_get_inode_info
cifs_get_fattr
smb2_query_path_info
smb2_compound_op
SMB2_open_init
smb2_reconnect
cifs_negotiate_protocol
smb2_negotiate
cifs_send_recv

    smb_send_rqst
    wait_for_response
                            cifs_demultiplex_thread
                              cifs_read_from_socket
                              cifs_readv_from_socket
                                server_unresponsive
                                cifs_reconnect
                                  __cifs_reconnect
                                  cifs_abort_connection
                                    mid->mid_state = MID_RETRY_NEEDED
                                    cifs_wake_up_task
    cifs_sync_mid_result
     // case MID_RETRY_NEEDED
     rc = -EAGAIN;

// In smb2_negotiate()
rc = -EHOSTDOWN;

The server_unresponsive() timeout triggers cifs_reconnect(), which aborts ongoing mid requests and causes the ls command to receive -EAGAIN, leading to -EHOSTDOWN.

Fix this by introducing a dedicated neg_start field to precisely tracks when the negotiate process begins. The timeout check now uses this accurate timestamp instead of lstrp, ensuring that:

1. Timeout is only triggered after negotiate has actually run for 20s
2. The mechanism doesn't interfere with concurrent recovery processes
3. Uninitialized timestamps (value 0) don't trigger false timeouts

Fixes 7ccc1465465d ("smb: client: fix hang in wait_for_response() for negproto")
---
 fs/smb/client/cifsglob.h | 1 +
 fs/smb/client/connect.c  | 7 ++++---
 2 files changed, 5 insertions(+), 3 deletions(-)
fs/smb/client/cifsglob.h +1 -0
index b89db44e0a8e3..eed29f043114a 100644
@@ -777,6 +777,7 @@ struct TCP_Server_Info {
 	__le32 session_key_id; /* retrieved from negotiate response and send in session setup request */
 	struct session_key session_key;
 	unsigned long lstrp; /* when we got last response from this server */
+	unsigned long neg_start; /* when negotiate started (jiffies) */
 	struct cifs_secmech secmech; /* crypto sec mech functs, descriptors */
 #define	CIFS_NEGFLAVOR_UNENCAP	1	/* wct == 17, but no ext_sec */
 #define	CIFS_NEGFLAVOR_EXTENDED	2	/* wct == 17, ext_sec bit set */
fs/smb/client/connect.c +4 -3
index e92c7b71626fd..b4b162c93106b 100644
@@ -679,12 +679,12 @@ server_unresponsive(struct TCP_Server_Info *server)
 	/*
 	 * If we're in the process of mounting a share or reconnecting a session
 	 * and the server abruptly shut down (e.g. socket wasn't closed, packet
-	 * had been ACK'ed but no SMB response), don't wait longer than 20s to
-	 * negotiate protocol.
+	 * had been ACK'ed but no SMB response), don't wait longer than 20s from
+	 * when negotiate actually started.
 	 */
 	spin_lock(&server->srv_lock);
 	if (server->tcpStatus == CifsInNegotiate &&
-	    time_after(jiffies, server->lstrp + 20 * HZ)) {
+	    time_after(jiffies, server->neg_start + 20 * HZ)) {
 		spin_unlock(&server->srv_lock);
 		cifs_reconnect(server, false);
 		return true;
@@ -4209,6 +4209,7 @@ cifs_negotiate_protocol(const unsigned int xid, struct cifs_ses *ses,
 
 	server->lstrp = jiffies;
 	server->tcpStatus = CifsInNegotiate;
+	server->neg_start = jiffies;
 	spin_unlock(&server->srv_lock);
 
 	rc = server->ops->negotiate(xid, ses, server);
Signature · 1 line
2.39.5
author maintainer
#118 lore raw

PATCH 118/178 smb: client: fix native SMB symlink traversal

In reply to Greg Kroah-Hartman

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

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

From: Paulo Alcantara <[email protected]>

[ Upstream commit 3363da82e02f1bddc54faa92ea430c6532e2cd2e ]

We've seen customers having shares mounted in paths like /??/C:/ or /??/UNC/foo.example.com/share in order to get their native SMB symlinks successfully followed from different mounts.

After commit 12b466eb52d9 ("cifs: Fix creating and resolving absolute NT-style symlinks"), the client would then convert absolute paths from "/??/C:/" to "/mnt/c/"
by default. The absolute paths would vary depending on the value of symlinkroot= mount option.

Fix this by restoring old behavior of not trying to convert absolute paths by default. Only do this if symlinkroot= was _explicitly_ set.

Before patch:

  $ mount.cifs //w22-fs0/test2 /mnt/1 -o vers=3.1.1,username=xxx,password=yyy
  $ ls -l /mnt/1/symlink2
  lrwxr-xr-x 1 root root 15 Jun 20 14:22 /mnt/1/symlink2 -> /mnt/c/testfile
  $ mkdir -p /??/C:; echo foo > //??/C:/testfile
  $ cat /mnt/1/symlink2
  cat: /mnt/1/symlink2: No such file or directory

After patch:

  $ mount.cifs //w22-fs0/test2 /mnt/1 -o vers=3.1.1,username=xxx,password=yyy
  $ ls -l /mnt/1/symlink2
  lrwxr-xr-x 1 root root 15 Jun 20 14:22 /mnt/1/symlink2 -> '/??/C:/testfile'
  $ mkdir -p /??/C:; echo foo > //??/C:/testfile
  $ cat /mnt/1/symlink2
  foo
Cc David Howells <[email protected]>
Cc Stefan Metzmacher <[email protected]>
Fixes 12b466eb52d9 ("cifs: Fix creating and resolving absolute NT-style symlinks")
---
 fs/smb/client/fs_context.c | 17 +++++++----------
 fs/smb/client/reparse.c    | 22 +++++++++++++---------
 2 files changed, 20 insertions(+), 19 deletions(-)
fs/smb/client/fs_context.c +7 -10
index a634a34d4086a..59ccc2229ab30 100644
@@ -1824,10 +1824,14 @@ static int smb3_fs_context_parse_param(struct fs_context *fc,
 			cifs_errorf(fc, "symlinkroot mount options must be absolute path\n");
 			goto cifs_parse_mount_err;
 		}
-		kfree(ctx->symlinkroot);
-		ctx->symlinkroot = kstrdup(param->string, GFP_KERNEL);
-		if (!ctx->symlinkroot)
+		if (strnlen(param->string, PATH_MAX) == PATH_MAX) {
+			cifs_errorf(fc, "symlinkroot path too long (max path length: %u)\n",
+				    PATH_MAX - 1);
 			goto cifs_parse_mount_err;
+		}
+		kfree(ctx->symlinkroot);
+		ctx->symlinkroot = param->string;
+		param->string = NULL;
 		break;
 	}
 	/* case Opt_ignore: - is ignored as expected ... */
@@ -1837,13 +1841,6 @@ static int smb3_fs_context_parse_param(struct fs_context *fc,
 		goto cifs_parse_mount_err;
 	}
 
-	/*
-	 * By default resolve all native absolute symlinks relative to "/mnt/".
-	 * Same default has drvfs driver running in WSL for resolving SMB shares.
-	 */
-	if (!ctx->symlinkroot)
-		ctx->symlinkroot = kstrdup("/mnt/", GFP_KERNEL);
-
 	return 0;
 
  cifs_parse_mount_err:
fs/smb/client/reparse.c +13 -9
index 1c40e42e4d897..5fa29a97ac154 100644
@@ -57,6 +57,7 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode,
 	struct reparse_symlink_data_buffer *buf = NULL;
 	struct cifs_open_info_data data = {};
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
+	const char *symroot = cifs_sb->ctx->symlinkroot;
 	struct inode *new;
 	struct kvec iov;
 	__le16 *path = NULL;
@@ -82,7 +83,8 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode,
 		.symlink_target = symlink_target,
 	};
 
-	if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') {
+	if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) &&
+	    symroot && symname[0] == '/') {
 		/*
 		 * This is a request to create an absolute symlink on the server
 		 * which does not support POSIX paths, and expects symlink in
@@ -92,7 +94,7 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode,
 		 * ensure compatibility of this symlink stored in absolute form
 		 * on the SMB server.
 		 */
-		if (!strstarts(symname, cifs_sb->ctx->symlinkroot)) {
+		if (!strstarts(symname, symroot)) {
 			/*
 			 * If the absolute Linux symlink target path is not
 			 * inside "symlinkroot" location then there is no way
@@ -101,12 +103,12 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode,
 			cifs_dbg(VFS,
 				 "absolute symlink '%s' cannot be converted to NT format "
 				 "because it is outside of symlinkroot='%s'\n",
-				 symname, cifs_sb->ctx->symlinkroot);
+				 symname, symroot);
 			rc = -EINVAL;
 			goto out;
 		}
-		len = strlen(cifs_sb->ctx->symlinkroot);
-		if (cifs_sb->ctx->symlinkroot[len-1] != '/')
+		len = strlen(symroot);
+		if (symroot[len - 1] != '/')
 			len++;
 		if (symname[len] >= 'a' && symname[len] <= 'z' &&
 		    (symname[len+1] == '/' || symname[len+1] == '\0')) {
@@ -782,6 +784,7 @@ int smb2_parse_native_symlink(char **target, const char *buf, unsigned int len,
 			      const char *full_path,
 			      struct cifs_sb_info *cifs_sb)
 {
+	const char *symroot = cifs_sb->ctx->symlinkroot;
 	char sep = CIFS_DIR_SEP(cifs_sb);
 	char *linux_target = NULL;
 	char *smb_target = NULL;
@@ -815,7 +818,8 @@ int smb2_parse_native_symlink(char **target, const char *buf, unsigned int len,
 		goto out;
 	}
 
-	if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && !relative) {
+	if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) &&
+	    symroot && !relative) {
 		/*
 		 * This is an absolute symlink from the server which does not
 		 * support POSIX paths, so the symlink is in NT-style path.
@@ -907,15 +911,15 @@ int smb2_parse_native_symlink(char **target, const char *buf, unsigned int len,
 		}
 
 		abs_path_len = strlen(abs_path)+1;
-		symlinkroot_len = strlen(cifs_sb->ctx->symlinkroot);
-		if (cifs_sb->ctx->symlinkroot[symlinkroot_len-1] == '/')
+		symlinkroot_len = strlen(symroot);
+		if (symroot[symlinkroot_len - 1] == '/')
 			symlinkroot_len--;
 		linux_target = kmalloc(symlinkroot_len + 1 + abs_path_len, GFP_KERNEL);
 		if (!linux_target) {
 			rc = -ENOMEM;
 			goto out;
 		}
-		memcpy(linux_target, cifs_sb->ctx->symlinkroot, symlinkroot_len);
+		memcpy(linux_target, symroot, symlinkroot_len);
 		linux_target[symlinkroot_len] = '/';
 		memcpy(linux_target + symlinkroot_len + 1, abs_path, abs_path_len);
 	} else if (smb_target[0] == sep && relative) {
Signature · 1 line
2.39.5
author maintainer
#119 lore raw

PATCH 119/178 netfs: Fix double put of request

In reply to Greg Kroah-Hartman

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

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

From: David Howells <[email protected]>

[ Upstream commit 9df7b5ebead649b00bf9a53a798e4bf83a1318fd ]

If a netfs request finishes during the pause loop, it will have the ref that belongs to the IN_PROGRESS flag removed at that point - however, if it then goes to the final wait loop, that will *also* put the ref because it sees that the IN_PROGRESS flag is clear and incorrectly assumes that this happened when it called the collector.

In fact, since IN_PROGRESS is clear, we shouldn't call the collector again since it's done all the cleanup, such as calling ->ki_complete().

Fix this by making netfs_collect_in_app() just return, indicating that we're done if IN_PROGRESS is removed.

---
 fs/netfs/misc.c              | 9 +++++++--
 include/trace/events/netfs.h | 1 +
 2 files changed, 8 insertions(+), 2 deletions(-)
fs/netfs/misc.c +7 -2
index d8b1a279dbda9..8b1c11ef32aa5 100644
@@ -383,6 +383,11 @@ static int netfs_collect_in_app(struct netfs_io_request *rreq,
 {
 	bool need_collect = false, inactive = true, done = true;
 
+	if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags)) {
+		trace_netfs_rreq(rreq, netfs_rreq_trace_recollect);
+		return 1; /* Done */
+	}
+
 	for (int i = 0; i < NR_IO_STREAMS; i++) {
 		struct netfs_io_subrequest *subreq;
 		struct netfs_io_stream *stream = &rreq->io_streams[i];
@@ -442,7 +447,7 @@ static ssize_t netfs_wait_for_in_progress(struct netfs_io_request *rreq,
 			case 1:
 				goto all_collected;
 			case 2:
-				if (!netfs_check_rreq_in_progress(rreq))
+				if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags))
 					break;
 				cond_resched();
 				continue;
@@ -512,7 +517,7 @@ static void netfs_wait_for_pause(struct netfs_io_request *rreq,
 			case 1:
 				goto all_collected;
 			case 2:
-				if (!netfs_check_rreq_in_progress(rreq) ||
+				if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags) ||
 				    !test_bit(NETFS_RREQ_PAUSE, &rreq->flags))
 					break;
 				cond_resched();
include/trace/events/netfs.h +1 -0
index 4175eec40048a..ecc1b852661e3 100644
@@ -56,6 +56,7 @@
 	EM(netfs_rreq_trace_dirty,		"DIRTY  ")	\
 	EM(netfs_rreq_trace_done,		"DONE   ")	\
 	EM(netfs_rreq_trace_free,		"FREE   ")	\
+	EM(netfs_rreq_trace_recollect,		"RECLLCT")	\
 	EM(netfs_rreq_trace_redirty,		"REDIRTY")	\
 	EM(netfs_rreq_trace_resubmit,		"RESUBMT")	\
 	EM(netfs_rreq_trace_set_abandon,	"S-ABNDN")	\
Signature · 1 line
2.39.5
author maintainer
#120 lore raw

PATCH 120/178 drm/xe: Allow dropping kunit dependency as built-in

In reply to Greg Kroah-Hartman

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

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

From: Harry Austen <[email protected]>

[ Upstream commit aa18d5769fcafe645a3ba01a9a69dde4f8dc8cc3 ]

Fix Kconfig symbol dependency on KUNIT, which isn't actually required for XE to be built-in. However, if KUNIT is enabled, it must be built-in too.

Fixes 08987a8b6820 ("drm/xe: Fix build with KUNIT=m")
Cc Lucas De Marchi <[email protected]>
Cc Thomas Hellström <[email protected]>
Cc Jani Nikula <[email protected]>
Cc Maarten Lankhorst <[email protected]>
(cherry picked from commit a559434880b320b83733d739733250815aecf1b0)
---
 drivers/gpu/drm/xe/Kconfig | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/xe/Kconfig +2 -1
index 9a46aafcb33ba..0dc3512876b3a 100644
@@ -1,7 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0-only
 config DRM_XE
 	tristate "Intel Xe Graphics"
-	depends on DRM && PCI && MMU && (m || (y && KUNIT=y))
+	depends on DRM && PCI && MMU
+	depends on KUNIT || !KUNIT
 	depends on INTEL_VSEC || !INTEL_VSEC
 	depends on X86_PLATFORM_DEVICES || !(X86 && ACPI)
 	select INTERVAL_TREE
Signature · 1 line
2.39.5
author maintainer
#121 lore raw

PATCH 121/178 x86/platform/amd: move final timeout check to after final sleep

In reply to Greg Kroah-Hartman

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

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

From: Jake Hillion <[email protected]>

[ Upstream commit f8afb12a2d7503de6558c23cacd7acbf6e9fe678 ]

__hsmp_send_message sleeps between result read attempts and has a timeout of 100ms. Under extreme load it's possible for these sleeps to take a long time, exceeding the 100ms. In this case the current code does not check the register and fails with ETIMEDOUT.

Refactor the loop to ensure there is at least one read of the register after a sleep of any duration. This removes instances of ETIMEDOUT with a single caller, even with a misbehaving scheduler. Tested on AMD Bergamo machines.

---
 drivers/platform/x86/amd/hsmp/hsmp.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/platform/x86/amd/hsmp/hsmp.c +5 -1
index a3ac09a90de45..ab877112f4c80 100644
@@ -99,7 +99,7 @@ static int __hsmp_send_message(struct hsmp_socket *sock, struct hsmp_message *ms
 	short_sleep = jiffies + msecs_to_jiffies(HSMP_SHORT_SLEEP);
 	timeout	= jiffies + msecs_to_jiffies(HSMP_MSG_TIMEOUT);
 
-	while (time_before(jiffies, timeout)) {
+	while (true) {
 		ret = sock->amd_hsmp_rdwr(sock, mbinfo->msg_resp_off, &mbox_status, HSMP_RD);
 		if (ret) {
 			dev_err(sock->dev, "Error %d reading mailbox status\n", ret);
@@ -108,6 +108,10 @@ static int __hsmp_send_message(struct hsmp_socket *sock, struct hsmp_message *ms
 
 		if (mbox_status != HSMP_STATUS_NOT_READY)
 			break;
+
+		if (!time_before(jiffies, timeout))
+			break;
+
 		if (time_before(jiffies, short_sleep))
 			usleep_range(50, 100);
 		else
Signature · 1 line
2.39.5
author maintainer
#104 lore raw

PATCH 104/178 drm/xe/guc: Enable w/a 16026508708

In reply to Greg Kroah-Hartman

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

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

From: John Harrison <[email protected]>

[ Upstream commit d3e8349edf7ed9eaf076ab3d9973331ccc20e26c ]

The workaround is only relevant to SRIOV but does affect all platforms.

Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487")

---
 drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 +
 drivers/gpu/drm/xe/xe_guc_ads.c       | 5 +++++
 drivers/gpu/drm/xe/xe_wa_oob.rules    | 2 ++
 3 files changed, 8 insertions(+)
drivers/gpu/drm/xe/abi/guc_klvs_abi.h +1 -0
index d633f1c739e43..7de8f827281fc 100644
@@ -367,6 +367,7 @@ enum xe_guc_klv_ids {
 	GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE			= 0x9008,
 	GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET				= 0x9009,
 	GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO					= 0x900a,
+	GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH					= 0x900b,
 };
 
 #endif
drivers/gpu/drm/xe/xe_guc_ads.c +5 -0
index 7031542a70ceb..1fccdef9e532e 100644
@@ -376,6 +376,11 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 					GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET,
 					&offset, &remain);
 
+	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_WA(gt, 16026508708))
+		guc_waklv_enable_simple(ads,
+					GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH,
+					&offset, &remain);
+
 	size = guc_ads_waklv_size(ads) - remain;
 	if (!size)
 		return;
drivers/gpu/drm/xe/xe_wa_oob.rules +2 -0
index 9b9e176992a83..9efc5accd43d1 100644
@@ -57,3 +57,5 @@ no_media_l3	MEDIA_VERSION(3000)
 		GRAPHICS_VERSION(1260), GRAPHICS_STEP(A0, B0)
 16023105232	GRAPHICS_VERSION_RANGE(2001, 3001)
 		MEDIA_VERSION_RANGE(1301, 3000)
+16026508708	GRAPHICS_VERSION_RANGE(1200, 3001)
+		MEDIA_VERSION_RANGE(1300, 3000)
Signature · 1 line
2.39.5
author maintainer
#122 lore raw

PATCH 122/178 drm/msm: Fix a fence leak in submit error path

In reply to Greg Kroah-Hartman

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

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

From: Rob Clark <[email protected]>

[ Upstream commit 5d319f75ccf7f0927425a7545aa1a22b3eedc189 ]

In error paths, we could unref the submit without calling drm_sched_entity_push_job(), so msm_job_free() will never get called. Since drm_sched_job_cleanup() will NULL out the s_fence, we can use that to detect this case.

Patchwork: https://patchwork.freedesktop.org/patch/653584/

---
 drivers/gpu/drm/msm/msm_gem_submit.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/gpu/drm/msm/msm_gem_submit.c +9 -0
index 3e9aa2cc38ef9..b2aeaecaa39b3 100644
@@ -85,6 +85,15 @@ void __msm_gem_submit_destroy(struct kref *kref)
 			container_of(kref, struct msm_gem_submit, ref);
 	unsigned i;
 
+	/*
+	 * In error paths, we could unref the submit without calling
+	 * drm_sched_entity_push_job(), so msm_job_free() will never
+	 * get called.  Since drm_sched_job_cleanup() will NULL out
+	 * s_fence, we can use that to detect this case.
+	 */
+	if (submit->base.s_fence)
+		drm_sched_job_cleanup(&submit->base);
+
 	if (submit->fence_id) {
 		spin_lock(&submit->queue->idr_lock);
 		idr_remove(&submit->queue->fence_idr, submit->fence_id);
Signature · 1 line
2.39.5
author maintainer
#123 lore raw

PATCH 123/178 drm/msm: Fix another leak in the submit error path

In reply to Greg Kroah-Hartman

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

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

From: Rob Clark <[email protected]>

[ Upstream commit f681c2aa8676a890eacc84044717ab0fd26e058f ]

put_unused_fd() doesn't free the installed file, if we've already done fd_install(). So we need to also free the sync_file.

Patchwork: https://patchwork.freedesktop.org/patch/653583/

---
 drivers/gpu/drm/msm/msm_gem_submit.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/gpu/drm/msm/msm_gem_submit.c +6 -2
index b2aeaecaa39b3..d4f71bb54e84c 100644
@@ -658,6 +658,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	struct msm_ringbuffer *ring;
 	struct msm_submit_post_dep *post_deps = NULL;
 	struct drm_syncobj **syncobjs_to_reset = NULL;
+	struct sync_file *sync_file = NULL;
 	int out_fence_fd = -1;
 	unsigned i;
 	int ret;
@@ -867,7 +868,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	}
 
 	if (ret == 0 && args->flags & MSM_SUBMIT_FENCE_FD_OUT) {
-		struct sync_file *sync_file = sync_file_create(submit->user_fence);
+		sync_file = sync_file_create(submit->user_fence);
 		if (!sync_file) {
 			ret = -ENOMEM;
 		} else {
@@ -901,8 +902,11 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 out_unlock:
 	mutex_unlock(&queue->lock);
 out_post_unlock:
-	if (ret && (out_fence_fd >= 0))
+	if (ret && (out_fence_fd >= 0)) {
 		put_unused_fd(out_fence_fd);
+		if (sync_file)
+			fput(sync_file->file);
+	}
 
 	if (!IS_ERR_OR_NULL(submit)) {
 		msm_gem_submit_put(submit);
Signature · 1 line
2.39.5
author maintainer
#124 lore raw

PATCH 124/178 ALSA: sb: Dont allow changing the DMA mode during operations

In reply to Greg Kroah-Hartman

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

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

From: Takashi Iwai <[email protected]>

[ Upstream commit ed29e073ba93f2d52832804cabdd831d5d357d33 ]

When a PCM stream is already running, one shouldn't change the DMA mode via kcontrol, which may screw up the hardware. Return -EBUSY instead.

---
 sound/isa/sb/sb16_main.c | 3 +++
 1 file changed, 3 insertions(+)
sound/isa/sb/sb16_main.c +3 -0
index 74db115250030..c4930efd44e3a 100644
@@ -703,6 +703,9 @@ static int snd_sb16_dma_control_put(struct snd_kcontrol *kcontrol, struct snd_ct
 	unsigned char nval, oval;
 	int change;
 	
+	if (chip->mode & (SB_MODE_PLAYBACK | SB_MODE_CAPTURE))
+		return -EBUSY;
+
 	nval = ucontrol->value.enumerated.item[0];
 	if (nval > 2)
 		return -EINVAL;
Signature · 1 line
2.39.5
author maintainer
#125 lore raw

PATCH 125/178 ALSA: sb: Force to disable DMAs once when DMA mode is changed

In reply to Greg Kroah-Hartman

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

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

From: Takashi Iwai <[email protected]>

[ Upstream commit 4c267ae2ef349639b4d9ebf00dd28586a82fdbe6 ]

When the DMA mode is changed on the (still real!) SB AWE32 after playing a stream and closing, the previous DMA setup was still silently kept, and it can confuse the hardware, resulting in the unexpected noises. As a workaround, enforce the disablement of DMA setups when the DMA setup is changed by the kcontrol.

https://bugzilla.kernel.org/show_bug.cgi?id=218185

---
 sound/isa/sb/sb16_main.c | 4 ++++
 1 file changed, 4 insertions(+)
sound/isa/sb/sb16_main.c +4 -0
index c4930efd44e3a..5a083eecaa6b9 100644
@@ -714,6 +714,10 @@ static int snd_sb16_dma_control_put(struct snd_kcontrol *kcontrol, struct snd_ct
 	change = nval != oval;
 	snd_sb16_set_dma_mode(chip, nval);
 	spin_unlock_irqrestore(&chip->reg_lock, flags);
+	if (change) {
+		snd_dma_disable(chip->dma8);
+		snd_dma_disable(chip->dma16);
+	}
 	return change;
 }
 
Signature · 1 line
2.39.5
author maintainer
#126 lore raw

PATCH 126/178 ata: libata-acpi: Do not assume 40 wire cable if no devices are enabled

In reply to Greg Kroah-Hartman

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

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

From: Tasos Sahanidis <[email protected]>

[ Upstream commit 33877220b8641b4cde474a4229ea92c0e3637883 ]

On at least an ASRock 990FX Extreme 4 with a VIA VT6330, the devices have not yet been enabled by the first time ata_acpi_cbl_80wire() is called. This means that the ata_for_each_dev loop is never entered, and a 40 wire cable is assumed.

The VIA controller on this board does not report the cable in the PCI config space, thus having to fall back to ACPI even though no SATA bridge is present.

The _GTM values are correctly reported by the firmware through ACPI, which has already set up faster transfer modes, but due to the above the controller is forced down to a maximum of UDMA/33.

Resolve this by modifying ata_acpi_cbl_80wire() to directly return the cable type. First, an unknown cable is assumed which preserves the mode set by the firmware, and then on subsequent calls when the devices have been enabled, an 80 wire cable is correctly detected.

Since the function now directly returns the cable type, it is renamed to ata_acpi_cbl_pata_type().

---
 drivers/ata/libata-acpi.c | 24 ++++++++++++++++--------
 drivers/ata/pata_via.c    |  6 ++----
 include/linux/libata.h    |  7 +++----
 3 files changed, 21 insertions(+), 16 deletions(-)
drivers/ata/libata-acpi.c +16 -8
index b7f0bf7955213..f2140fc06ba0f 100644
@@ -514,15 +514,19 @@ unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev,
 EXPORT_SYMBOL_GPL(ata_acpi_gtm_xfermask);
 
 /**
- * ata_acpi_cbl_80wire		-	Check for 80 wire cable
+ * ata_acpi_cbl_pata_type - Return PATA cable type
  * @ap: Port to check
- * @gtm: GTM data to use
  *
- * Return 1 if the @gtm indicates the BIOS selected an 80wire mode.
+ * Return ATA_CBL_PATA* according to the transfer mode selected by BIOS
  */
-int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm)
+int ata_acpi_cbl_pata_type(struct ata_port *ap)
 {
 	struct ata_device *dev;
+	int ret = ATA_CBL_PATA_UNK;
+	const struct ata_acpi_gtm *gtm = ata_acpi_init_gtm(ap);
+
+	if (!gtm)
+		return ATA_CBL_PATA40;
 
 	ata_for_each_dev(dev, &ap->link, ENABLED) {
 		unsigned int xfer_mask, udma_mask;
@@ -530,13 +534,17 @@ int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm)
 		xfer_mask = ata_acpi_gtm_xfermask(dev, gtm);
 		ata_unpack_xfermask(xfer_mask, NULL, NULL, &udma_mask);
 
-		if (udma_mask & ~ATA_UDMA_MASK_40C)
-			return 1;
+		ret = ATA_CBL_PATA40;
+
+		if (udma_mask & ~ATA_UDMA_MASK_40C) {
+			ret = ATA_CBL_PATA80;
+			break;
+		}
 	}
 
-	return 0;
+	return ret;
 }
-EXPORT_SYMBOL_GPL(ata_acpi_cbl_80wire);
+EXPORT_SYMBOL_GPL(ata_acpi_cbl_pata_type);
 
 static void ata_acpi_gtf_to_tf(struct ata_device *dev,
 			       const struct ata_acpi_gtf *gtf,
drivers/ata/pata_via.c +2 -4
index d82728a01832b..bb80e7800dcbe 100644
@@ -201,11 +201,9 @@ static int via_cable_detect(struct ata_port *ap) {
 	   two drives */
 	if (ata66 & (0x10100000 >> (16 * ap->port_no)))
 		return ATA_CBL_PATA80;
+
 	/* Check with ACPI so we can spot BIOS reported SATA bridges */
-	if (ata_acpi_init_gtm(ap) &&
-	    ata_acpi_cbl_80wire(ap, ata_acpi_init_gtm(ap)))
-		return ATA_CBL_PATA80;
-	return ATA_CBL_PATA40;
+	return ata_acpi_cbl_pata_type(ap);
 }
 
 static int via_pre_reset(struct ata_link *link, unsigned long deadline)
include/linux/libata.h +3 -4
index e5695998acb02..ec3b0c9c2a8cf 100644
@@ -1363,7 +1363,7 @@ int ata_acpi_stm(struct ata_port *ap, const struct ata_acpi_gtm *stm);
 int ata_acpi_gtm(struct ata_port *ap, struct ata_acpi_gtm *stm);
 unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev,
 				   const struct ata_acpi_gtm *gtm);
-int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm);
+int ata_acpi_cbl_pata_type(struct ata_port *ap);
 #else
 static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 {
@@ -1388,10 +1388,9 @@ static inline unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev,
 	return 0;
 }
 
-static inline int ata_acpi_cbl_80wire(struct ata_port *ap,
-				      const struct ata_acpi_gtm *gtm)
+static inline int ata_acpi_cbl_pata_type(struct ata_port *ap)
 {
-	return 0;
+	return ATA_CBL_PATA40;
 }
 #endif
 
Signature · 1 line
2.39.5
author maintainer
#127 lore raw

PATCH 127/178 ata: pata_cs5536: fix build on 32-bit UML

In reply to Greg Kroah-Hartman

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

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

From: Johannes Berg <[email protected]>

[ Upstream commit fe5b391fc56f77cf3c22a9dd4f0ce20db0e3533f ]

On 32-bit ARCH=um, CONFIG_X86_32 is still defined, so it doesn't indicate building on real X86 machines. There's no MSR on UML though, so add a check for CONFIG_X86.

---
 drivers/ata/pata_cs5536.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ata/pata_cs5536.c +1 -1
index b811efd2cc346..73e81e160c91f 100644
@@ -27,7 +27,7 @@
 #include <scsi/scsi_host.h>
 #include <linux/dmi.h>
 
-#ifdef CONFIG_X86_32
+#if defined(CONFIG_X86) && defined(CONFIG_X86_32)
 #include <asm/msr.h>
 static int use_msr;
 module_param_named(msr, use_msr, int, 0644);
Signature · 1 line
2.39.5
author maintainer
#128 lore raw

PATCH 128/178 ASoC: amd: yc: Add quirk for MSI Bravo 17 D7VF internal mic

In reply to Greg Kroah-Hartman

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

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

From: Gabriel Santese <[email protected]>

[ Upstream commit ba06528ad5a31923efc24324706116ccd17e12d8 ]

MSI Bravo 17 (D7VF), like other laptops from the family, has broken ACPI tables and needs a quirk for internal mic to work properly.

---
 sound/soc/amd/yc/acp6x-mach.c | 7 +++++++
 1 file changed, 7 insertions(+)
sound/soc/amd/yc/acp6x-mach.c +7 -0
index b27966f82c8b6..4f8481c6802b1 100644
@@ -451,6 +451,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "Bravo 17 D7VEK"),
 		}
 	},
+	{
+		.driver_data = &acp6x_card,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Micro-Star International Co., Ltd."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Bravo 17 D7VF"),
+		}
+	},
 	{
 		.driver_data = &acp6x_card,
 		.matches = {
Signature · 1 line
2.39.5
author maintainer
#129 lore raw

PATCH 129/178 platform/x86/amd/pmc: Add PCSpecialist Lafite Pro V 14M to 8042 quirks list

In reply to Greg Kroah-Hartman

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

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

From: Mario Limonciello <[email protected]>

[ Upstream commit 9ba75ccad85708c5a484637dccc1fc59295b0a83 ]

Every other s2idle cycle fails to reach hardware sleep when keyboard wakeup is enabled. This appears to be an EC bug, but the vendor refuses to fix it.

It was confirmed that turning off i8042 wakeup avoids ths issue (albeit keyboard wakeup is disabled). Take the lesser of two evils and add it to the i8042 quirk list.

---
 drivers/platform/x86/amd/pmc/pmc-quirks.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/platform/x86/amd/pmc/pmc-quirks.c +9 -0
index 2e3f6fc67c568..7ed12c1d3b34c 100644
@@ -224,6 +224,15 @@ static const struct dmi_system_id fwbug_list[] = {
 			DMI_MATCH(DMI_BOARD_NAME, "WUJIE14-GX4HRXL"),
 		}
 	},
+	/* https://bugzilla.kernel.org/show_bug.cgi?id=220116 */
+	{
+		.ident = "PCSpecialist Lafite Pro V 14M",
+		.driver_data = &quirk_spurious_8042,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "PCSpecialist"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Lafite Pro V 14M"),
+		}
+	},
 	{}
 };
 
Signature · 1 line
2.39.5
author maintainer
#130 lore raw

PATCH 130/178 genirq/irq_sim: Initialize work context pointers properly

In reply to Greg Kroah-Hartman

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

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

From: Gyeyoung Baek <[email protected]>

[ Upstream commit 8a2277a3c9e4cc5398f80821afe7ecbe9bdf2819 ]

Initialize ops member's pointers properly by using kzalloc() instead of kmalloc() when allocating the simulation work context. Otherwise the pointers contain random content leading to invalid dereferencing.

---
 kernel/irq/irq_sim.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/irq/irq_sim.c +1 -1
index 1a3d483548e2f..ae4c9cbd1b4b9 100644
@@ -202,7 +202,7 @@ struct irq_domain *irq_domain_create_sim_full(struct fwnode_handle *fwnode,
 					      void *data)
 {
 	struct irq_sim_work_ctx *work_ctx __free(kfree) =
-				kmalloc(sizeof(*work_ctx), GFP_KERNEL);
+				kzalloc(sizeof(*work_ctx), GFP_KERNEL);
 
 	if (!work_ctx)
 		return ERR_PTR(-ENOMEM);
Signature · 1 line
2.39.5
author maintainer
#131 lore raw

PATCH 131/178 powerpc: Fix struct termio related ioctl macros

In reply to Greg Kroah-Hartman

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

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

From: Madhavan Srinivasan <[email protected]>

[ Upstream commit ab107276607af90b13a5994997e19b7b9731e251 ]

Since termio interface is now obsolete, include/uapi/asm/ioctls.h has some constant macros referring to "struct termio", this caused build failure at userspace.

In file included from /usr/include/asm/ioctl.h:12,
from /usr/include/asm/ioctls.h:5,
from tst-ioctls.c:3:
tst-ioctls.c: In function 'get_TCGETA':
tst-ioctls.c:12:10: error: invalid application of 'sizeof' to incomplete type 'struct termio'

   12 |   return TCGETA;
      |          ^~~~~~

Even though termios.h provides "struct termio", trying to juggle definitions around to make it compile could introduce regressions. So better to open code it.

---
 arch/powerpc/include/uapi/asm/ioctls.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
arch/powerpc/include/uapi/asm/ioctls.h +4 -4
index 2c145da3b774a..b5211e413829a 100644
@@ -23,10 +23,10 @@
 #define TCSETSW		_IOW('t', 21, struct termios)
 #define TCSETSF		_IOW('t', 22, struct termios)
 
-#define TCGETA		_IOR('t', 23, struct termio)
-#define TCSETA		_IOW('t', 24, struct termio)
-#define TCSETAW		_IOW('t', 25, struct termio)
-#define TCSETAF		_IOW('t', 28, struct termio)
+#define TCGETA		0x40147417 /* _IOR('t', 23, struct termio) */
+#define TCSETA		0x80147418 /* _IOW('t', 24, struct termio) */
+#define TCSETAW		0x80147419 /* _IOW('t', 25, struct termio) */
+#define TCSETAF		0x8014741c /* _IOW('t', 28, struct termio) */
 
 #define TCSBRK		_IO('t', 29)
 #define TCXONC		_IO('t', 30)
Signature · 1 line
2.39.5
author maintainer
#105 lore raw

PATCH 105/178 drm/xe/guc_pc: Add _locked variant for min/max freq

In reply to Greg Kroah-Hartman

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

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

From: Lucas De Marchi <[email protected]>

[ Upstream commit d8390768dcf6f5a78af56aa03797a076871b01f3 ]

There are places in which the getters/setters are called one after the other causing a multiple lock()/unlock(). These are not currently a problem since they are all happening from the same thread, but there's a race possibility as calls are added outside of the early init when the max/min and stashed values need to be correlated.

Add the _locked() variants to prepare for that.

(cherry picked from commit 1beae9aa2b88d3a02eb666e7b777eb2d7bc645f4)

Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487")

---
 drivers/gpu/drm/xe/xe_guc_pc.c | 123 ++++++++++++++++++---------------
 1 file changed, 69 insertions(+), 54 deletions(-)
drivers/gpu/drm/xe/xe_guc_pc.c +69 -54
index a7b8bacfe64ef..23a4c525c03bf 100644
@@ -5,6 +5,7 @@
 
 #include "xe_guc_pc.h"
 
+#include <linux/cleanup.h>
 #include <linux/delay.h>
 #include <linux/ktime.h>
 
@@ -538,6 +539,25 @@ u32 xe_guc_pc_get_rpn_freq(struct xe_guc_pc *pc)
 	return pc->rpn_freq;
 }
 
+static int xe_guc_pc_get_min_freq_locked(struct xe_guc_pc *pc, u32 *freq)
+{
+	int ret;
+
+	lockdep_assert_held(&pc->freq_lock);
+
+	/* Might be in the middle of a gt reset */
+	if (!pc->freq_ready)
+		return -EAGAIN;
+
+	ret = pc_action_query_task_state(pc);
+	if (ret)
+		return ret;
+
+	*freq = pc_get_min_freq(pc);
+
+	return 0;
+}
+
 /**
  * xe_guc_pc_get_min_freq - Get the min operational frequency
  * @pc: The GuC PC
@@ -547,27 +567,29 @@ u32 xe_guc_pc_get_rpn_freq(struct xe_guc_pc *pc)
  *         -EAGAIN if GuC PC not ready (likely in middle of a reset).
  */
 int xe_guc_pc_get_min_freq(struct xe_guc_pc *pc, u32 *freq)
+{
+	guard(mutex)(&pc->freq_lock);
+
+	return xe_guc_pc_get_min_freq_locked(pc, freq);
+}
+
+static int xe_guc_pc_set_min_freq_locked(struct xe_guc_pc *pc, u32 freq)
 {
 	int ret;
 
-	xe_device_assert_mem_access(pc_to_xe(pc));
+	lockdep_assert_held(&pc->freq_lock);
 
-	mutex_lock(&pc->freq_lock);
-	if (!pc->freq_ready) {
-		/* Might be in the middle of a gt reset */
-		ret = -EAGAIN;
-		goto out;
-	}
+	/* Might be in the middle of a gt reset */
+	if (!pc->freq_ready)
+		return -EAGAIN;
 
-	ret = pc_action_query_task_state(pc);
+	ret = pc_set_min_freq(pc, freq);
 	if (ret)
-		goto out;
+		return ret;
 
-	*freq = pc_get_min_freq(pc);
+	pc->user_requested_min = freq;
 
-out:
-	mutex_unlock(&pc->freq_lock);
-	return ret;
+	return 0;
 }
 
 /**
@@ -580,25 +602,29 @@ int xe_guc_pc_get_min_freq(struct xe_guc_pc *pc, u32 *freq)
  *         -EINVAL if value out of bounds.
  */
 int xe_guc_pc_set_min_freq(struct xe_guc_pc *pc, u32 freq)
+{
+	guard(mutex)(&pc->freq_lock);
+
+	return xe_guc_pc_set_min_freq_locked(pc, freq);
+}
+
+static int xe_guc_pc_get_max_freq_locked(struct xe_guc_pc *pc, u32 *freq)
 {
 	int ret;
 
-	mutex_lock(&pc->freq_lock);
-	if (!pc->freq_ready) {
-		/* Might be in the middle of a gt reset */
-		ret = -EAGAIN;
-		goto out;
-	}
+	lockdep_assert_held(&pc->freq_lock);
 
-	ret = pc_set_min_freq(pc, freq);
+	/* Might be in the middle of a gt reset */
+	if (!pc->freq_ready)
+		return -EAGAIN;
+
+	ret = pc_action_query_task_state(pc);
 	if (ret)
-		goto out;
+		return ret;
 
-	pc->user_requested_min = freq;
+	*freq = pc_get_max_freq(pc);
 
-out:
-	mutex_unlock(&pc->freq_lock);
-	return ret;
+	return 0;
 }
 
 /**
@@ -610,25 +636,29 @@ int xe_guc_pc_set_min_freq(struct xe_guc_pc *pc, u32 freq)
  *         -EAGAIN if GuC PC not ready (likely in middle of a reset).
  */
 int xe_guc_pc_get_max_freq(struct xe_guc_pc *pc, u32 *freq)
+{
+	guard(mutex)(&pc->freq_lock);
+
+	return xe_guc_pc_get_max_freq_locked(pc, freq);
+}
+
+static int xe_guc_pc_set_max_freq_locked(struct xe_guc_pc *pc, u32 freq)
 {
 	int ret;
 
-	mutex_lock(&pc->freq_lock);
-	if (!pc->freq_ready) {
-		/* Might be in the middle of a gt reset */
-		ret = -EAGAIN;
-		goto out;
-	}
+	lockdep_assert_held(&pc->freq_lock);
 
-	ret = pc_action_query_task_state(pc);
+	/* Might be in the middle of a gt reset */
+	if (!pc->freq_ready)
+		return -EAGAIN;
+
+	ret = pc_set_max_freq(pc, freq);
 	if (ret)
-		goto out;
+		return ret;
 
-	*freq = pc_get_max_freq(pc);
+	pc->user_requested_max = freq;
 
-out:
-	mutex_unlock(&pc->freq_lock);
-	return ret;
+	return 0;
 }
 
 /**
@@ -642,24 +672,9 @@ int xe_guc_pc_get_max_freq(struct xe_guc_pc *pc, u32 *freq)
  */
 int xe_guc_pc_set_max_freq(struct xe_guc_pc *pc, u32 freq)
 {
-	int ret;
-
-	mutex_lock(&pc->freq_lock);
-	if (!pc->freq_ready) {
-		/* Might be in the middle of a gt reset */
-		ret = -EAGAIN;
-		goto out;
-	}
-
-	ret = pc_set_max_freq(pc, freq);
-	if (ret)
-		goto out;
+	guard(mutex)(&pc->freq_lock);
 
-	pc->user_requested_max = freq;
-
-out:
-	mutex_unlock(&pc->freq_lock);
-	return ret;
+	return xe_guc_pc_set_max_freq_locked(pc, freq);
 }
 
 /**
Signature · 1 line
2.39.5
author maintainer
#132 lore raw

PATCH 132/178 ASoC: amd: yc: update quirk data for HP Victus

In reply to Greg Kroah-Hartman

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

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

From: Raven Black <[email protected]>

[ Upstream commit 13b86ea92ebf0fa587fbadfb8a60ca2e9993203f ]

Make the internal microphone work on HP Victus laptops.

---
 sound/soc/amd/yc/acp6x-mach.c | 7 +++++++
 1 file changed, 7 insertions(+)
sound/soc/amd/yc/acp6x-mach.c +7 -0
index 4f8481c6802b1..723cb7bc12851 100644
@@ -521,6 +521,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "OMEN by HP Gaming Laptop 16z-n000"),
 		}
 	},
+	{
+		.driver_data = &acp6x_card,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "HP"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Victus by HP Gaming Laptop 15-fb2xxx"),
+		}
+	},
 	{
 		.driver_data = &acp6x_card,
 		.matches = {
Signature · 1 line
2.39.5
author maintainer
#133 lore raw

PATCH 133/178 regulator: fan53555: add enable_time support and soft-start times

In reply to Greg Kroah-Hartman

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

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

From: Heiko Stuebner <[email protected]>

[ Upstream commit 8acfb165a492251a08a22a4fa6497a131e8c2609 ]

The datasheets for all the fan53555 variants (and clones using the same interface) define so called soft start times, from enabling the regulator until at least some percentage of the output (i.e. 92% for the rk860x types) are available.

The regulator framework supports this with the enable_time property but currently the fan53555 driver does not define enable_times for any variant.

I ran into a problem with this while testing the new driver for the Rockchip NPUs (rocket), which does runtime-pm including disabling and enabling a rk8602 as needed. When reenabling the regulator while running a load, fatal hangs could be observed while enabling the associated power-domain, which the regulator supplies.

Experimentally setting the regulator to always-on, made the issue disappear, leading to the missing delay to let power stabilize. And as expected, setting the enable-time to a non-zero value according to the datasheet also resolved the regulator-issue.

The datasheets in nearly all cases only specify "typical" values, except for the fan53555 type 08. There both a typical and maximum value are listed - 40uS apart.

For all typical values I've added 100uS to be on the safe side. Individual details for the relevant regulators below:

- fan53526:
The datasheet for all variants lists a typical value of 150uS, so make that 250uS with safety margin.
- fan53555:
types 08 and 18 (unsupported) are given a typical enable time of 135uS but also a maximum of 175uS so use that value. All the other types only have a typical time in the datasheet of 300uS, so give a bit margin by setting it to 400uS.
- rk8600 + rk8602:
Datasheet reports a typical value of 260us, so use 360uS to be safe.
- syr82x + syr83x:
All datasheets report typical soft-start values of 300uS for these regulators, so use 400uS.
- tcs452x:
Datasheet sadly does not report a soft-start time, so I've not set an enable-time

---
 drivers/regulator/fan53555.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
drivers/regulator/fan53555.c +14 -0
index bd9447dac5967..c282236959b18 100644
@@ -147,6 +147,7 @@ struct fan53555_device_info {
 	unsigned int slew_mask;
 	const unsigned int *ramp_delay_table;
 	unsigned int n_ramp_values;
+	unsigned int enable_time;
 	unsigned int slew_rate;
 };
 
@@ -282,6 +283,7 @@ static int fan53526_voltages_setup_fairchild(struct fan53555_device_info *di)
 	di->slew_mask = CTL_SLEW_MASK;
 	di->ramp_delay_table = slew_rates;
 	di->n_ramp_values = ARRAY_SIZE(slew_rates);
+	di->enable_time = 250;
 	di->vsel_count = FAN53526_NVOLTAGES;
 
 	return 0;
@@ -296,10 +298,12 @@ static int fan53555_voltages_setup_fairchild(struct fan53555_device_info *di)
 		case FAN53555_CHIP_REV_00:
 			di->vsel_min = 600000;
 			di->vsel_step = 10000;
+			di->enable_time = 400;
 			break;
 		case FAN53555_CHIP_REV_13:
 			di->vsel_min = 800000;
 			di->vsel_step = 10000;
+			di->enable_time = 400;
 			break;
 		default:
 			dev_err(di->dev,
@@ -311,13 +315,19 @@ static int fan53555_voltages_setup_fairchild(struct fan53555_device_info *di)
 	case FAN53555_CHIP_ID_01:
 	case FAN53555_CHIP_ID_03:
 	case FAN53555_CHIP_ID_05:
+		di->vsel_min = 600000;
+		di->vsel_step = 10000;
+		di->enable_time = 400;
+		break;
 	case FAN53555_CHIP_ID_08:
 		di->vsel_min = 600000;
 		di->vsel_step = 10000;
+		di->enable_time = 175;
 		break;
 	case FAN53555_CHIP_ID_04:
 		di->vsel_min = 603000;
 		di->vsel_step = 12826;
+		di->enable_time = 400;
 		break;
 	default:
 		dev_err(di->dev,
@@ -350,6 +360,7 @@ static int fan53555_voltages_setup_rockchip(struct fan53555_device_info *di)
 	di->slew_mask = CTL_SLEW_MASK;
 	di->ramp_delay_table = slew_rates;
 	di->n_ramp_values = ARRAY_SIZE(slew_rates);
+	di->enable_time = 360;
 	di->vsel_count = FAN53555_NVOLTAGES;
 
 	return 0;
@@ -372,6 +383,7 @@ static int rk8602_voltages_setup_rockchip(struct fan53555_device_info *di)
 	di->slew_mask = CTL_SLEW_MASK;
 	di->ramp_delay_table = slew_rates;
 	di->n_ramp_values = ARRAY_SIZE(slew_rates);
+	di->enable_time = 360;
 	di->vsel_count = RK8602_NVOLTAGES;
 
 	return 0;
@@ -395,6 +407,7 @@ static int fan53555_voltages_setup_silergy(struct fan53555_device_info *di)
 	di->slew_mask = CTL_SLEW_MASK;
 	di->ramp_delay_table = slew_rates;
 	di->n_ramp_values = ARRAY_SIZE(slew_rates);
+	di->enable_time = 400;
 	di->vsel_count = FAN53555_NVOLTAGES;
 
 	return 0;
@@ -594,6 +607,7 @@ static int fan53555_regulator_register(struct fan53555_device_info *di,
 	rdesc->ramp_mask = di->slew_mask;
 	rdesc->ramp_delay_table = di->ramp_delay_table;
 	rdesc->n_ramp_values = di->n_ramp_values;
+	rdesc->enable_time = di->enable_time;
 	rdesc->owner = THIS_MODULE;
 
 	rdev = devm_regulator_register(di->dev, &di->desc, config);
Signature · 1 line
2.39.5
author maintainer
#134 lore raw

PATCH 134/178 scsi: target: Fix NULL pointer dereference in core_scsi3_decode_spec_i_port()

In reply to Greg Kroah-Hartman

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

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

From: Maurizio Lombardi <[email protected]>

[ Upstream commit d8ab68bdb294b09a761e967dad374f2965e1913f ]

The function core_scsi3_decode_spec_i_port(), in its error code path, unconditionally calls core_scsi3_lunacl_undepend_item() passing the dest_se_deve pointer, which may be NULL.

This can lead to a NULL pointer dereference if dest_se_deve remains unset.

SPC-3 PR SPEC_I_PT: Unable to locate dest_tpg

Unable to handle kernel paging request at virtual address dfff800000000012
Call trace:
  core_scsi3_lunacl_undepend_item+0x2c/0xf0 [target_core_mod] (P)
  core_scsi3_decode_spec_i_port+0x120c/0x1c30 [target_core_mod]
  core_scsi3_emulate_pro_register+0x6b8/0xcd8 [target_core_mod]
  target_scsi3_emulate_pr_out+0x56c/0x840 [target_core_mod]

Fix this by adding a NULL check before calling
core_scsi3_lunacl_undepend_item()

---
 drivers/target/target_core_pr.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/target/target_core_pr.c +3 -1
index 34cf2c399b399..70905805cb175 100644
@@ -1842,7 +1842,9 @@ core_scsi3_decode_spec_i_port(
 		}
 
 		kmem_cache_free(t10_pr_reg_cache, dest_pr_reg);
-		core_scsi3_lunacl_undepend_item(dest_se_deve);
+
+		if (dest_se_deve)
+			core_scsi3_lunacl_undepend_item(dest_se_deve);
 
 		if (is_local)
 			continue;
Signature · 1 line
2.39.5
author maintainer
#106 lore raw

PATCH 106/178 drm/xe: Split xe_device_td_flush()

In reply to Greg Kroah-Hartman

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

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

From: Lucas De Marchi <[email protected]>

[ Upstream commit a1eec6cae95a1a0888cb8370338822ca81cd9436 ]

xe_device_td_flush() has 2 possible implementations: an entire L2 flush or a transient flush, depending on WA 16023588340. Make this clear by splitting the function so it calls each of them.

(cherry picked from commit 5e300ed8a545bdffc26b579c526b5fef7b2d5365)

Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487")

---
 drivers/gpu/drm/xe/xe_device.c | 68 ++++++++++++++++++++--------------
 1 file changed, 40 insertions(+), 28 deletions(-)
drivers/gpu/drm/xe/xe_device.c +40 -28
index 00191227bc95c..38fdddd7262aa 100644
@@ -972,38 +972,15 @@ void xe_device_wmb(struct xe_device *xe)
 		xe_mmio_write32(xe_root_tile_mmio(xe), VF_CAP_REG, 0);
 }
 
-/**
- * xe_device_td_flush() - Flush transient L3 cache entries
- * @xe: The device
- *
- * Display engine has direct access to memory and is never coherent with L3/L4
- * caches (or CPU caches), however KMD is responsible for specifically flushing
- * transient L3 GPU cache entries prior to the flip sequence to ensure scanout
- * can happen from such a surface without seeing corruption.
- *
- * Display surfaces can be tagged as transient by mapping it using one of the
- * various L3:XD PAT index modes on Xe2.
- *
- * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed
- * at the end of each submission via PIPE_CONTROL for compute/render, since SA
- * Media is not coherent with L3 and we want to support render-vs-media
- * usescases. For other engines like copy/blt the HW internally forces uncached
- * behaviour, hence why we can skip the TDF on such platforms.
+/*
+ * Issue a TRANSIENT_FLUSH_REQUEST and wait for completion on each gt.
  */
-void xe_device_td_flush(struct xe_device *xe)
+static void tdf_request_sync(struct xe_device *xe)
 {
-	struct xe_gt *gt;
 	unsigned int fw_ref;
+	struct xe_gt *gt;
 	u8 id;
 
-	if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20)
-		return;
-
-	if (XE_WA(xe_root_mmio_gt(xe), 16023588340)) {
-		xe_device_l2_flush(xe);
-		return;
-	}
-
 	for_each_gt(gt, xe, id) {
 		if (xe_gt_is_media_type(gt))
 			continue;
@@ -1013,6 +990,7 @@ void xe_device_td_flush(struct xe_device *xe)
 			return;
 
 		xe_mmio_write32(&gt->mmio, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST);
+
 		/*
 		 * FIXME: We can likely do better here with our choice of
 		 * timeout. Currently we just assume the worst case, i.e. 150us,
@@ -1043,15 +1021,49 @@ void xe_device_l2_flush(struct xe_device *xe)
 		return;
 
 	spin_lock(&gt->global_invl_lock);
-	xe_mmio_write32(&gt->mmio, XE2_GLOBAL_INVAL, 0x1);
 
+	xe_mmio_write32(&gt->mmio, XE2_GLOBAL_INVAL, 0x1);
 	if (xe_mmio_wait32(&gt->mmio, XE2_GLOBAL_INVAL, 0x1, 0x0, 500, NULL, true))
 		xe_gt_err_once(gt, "Global invalidation timeout\n");
+
 	spin_unlock(&gt->global_invl_lock);
 
 	xe_force_wake_put(gt_to_fw(gt), fw_ref);
 }
 
+/**
+ * xe_device_td_flush() - Flush transient L3 cache entries
+ * @xe: The device
+ *
+ * Display engine has direct access to memory and is never coherent with L3/L4
+ * caches (or CPU caches), however KMD is responsible for specifically flushing
+ * transient L3 GPU cache entries prior to the flip sequence to ensure scanout
+ * can happen from such a surface without seeing corruption.
+ *
+ * Display surfaces can be tagged as transient by mapping it using one of the
+ * various L3:XD PAT index modes on Xe2.
+ *
+ * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed
+ * at the end of each submission via PIPE_CONTROL for compute/render, since SA
+ * Media is not coherent with L3 and we want to support render-vs-media
+ * usescases. For other engines like copy/blt the HW internally forces uncached
+ * behaviour, hence why we can skip the TDF on such platforms.
+ */
+void xe_device_td_flush(struct xe_device *xe)
+{
+	struct xe_gt *root_gt;
+
+	if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20)
+		return;
+
+	root_gt = xe_root_mmio_gt(xe);
+	if (XE_WA(root_gt, 16023588340))
+		/* A transient flush is not sufficient: flush the L2 */
+		xe_device_l2_flush(xe);
+	else
+		tdf_request_sync(xe);
+}
+
 u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size)
 {
 	return xe_device_has_flat_ccs(xe) ?
Signature · 1 line
2.39.5
author maintainer
#107 lore raw

PATCH 107/178 drm/xe/bmg: Update Wa_14022085890

In reply to Greg Kroah-Hartman

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

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

From: Vinay Belgaumkar <[email protected]>

[ Upstream commit a5c7dcdd969f2248cc91d65e5ac852859fc8dac2 ]

Set GT min frequency to 1200Mhz once driver load is complete.

v2: Review comments (Rodrigo)
v3: Apply Wa earlier so user_req_min is not clobbered.
v4: Apply to all GTs (Lucas)

Cc Matt Roper <[email protected]>
Cc Rodrigo Vivi <[email protected]>
(cherry picked from commit bdde16c9ac5cb56ad2ee19792222fa1853577af7)

Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487")

---
 drivers/gpu/drm/xe/xe_guc_pc.c     | 5 +++++
 drivers/gpu/drm/xe/xe_wa_oob.rules | 4 ++++
 2 files changed, 9 insertions(+)
drivers/gpu/drm/xe/xe_guc_pc.c +5 -0
index 23a4c525c03bf..28b97a2c14e3b 100644
@@ -52,6 +52,7 @@
 
 #define LNL_MERT_FREQ_CAP	800
 #define BMG_MERT_FREQ_CAP	2133
+#define BMG_MIN_FREQ		1200
 
 #define SLPC_RESET_TIMEOUT_MS 5 /* roughly 5ms, but no need for precision */
 #define SLPC_RESET_EXTENDED_TIMEOUT_MS 1000 /* To be used only at pc_start */
@@ -817,6 +818,7 @@ void xe_guc_pc_init_early(struct xe_guc_pc *pc)
 
 static int pc_adjust_freq_bounds(struct xe_guc_pc *pc)
 {
+	struct xe_tile *tile = gt_to_tile(pc_to_gt(pc));
 	int ret;
 
 	lockdep_assert_held(&pc->freq_lock);
@@ -843,6 +845,9 @@ static int pc_adjust_freq_bounds(struct xe_guc_pc *pc)
 	if (pc_get_min_freq(pc) > pc->rp0_freq)
 		ret = pc_set_min_freq(pc, pc->rp0_freq);
 
+	if (XE_WA(tile->primary_gt, 14022085890))
+		ret = pc_set_min_freq(pc, max(BMG_MIN_FREQ, pc_get_min_freq(pc)));
+
 out:
 	return ret;
 }
drivers/gpu/drm/xe/xe_wa_oob.rules +4 -0
index 9efc5accd43d1..320766f6c5dff 100644
@@ -59,3 +59,7 @@ no_media_l3	MEDIA_VERSION(3000)
 		MEDIA_VERSION_RANGE(1301, 3000)
 16026508708	GRAPHICS_VERSION_RANGE(1200, 3001)
 		MEDIA_VERSION_RANGE(1300, 3000)
+
+# SoC workaround - currently applies to all platforms with the following
+# primary GT GMDID
+14022085890	GRAPHICS_VERSION(2001)
Signature · 1 line
2.39.5
author maintainer
#108 lore raw

PATCH 108/178 drm/xe/bmg: Update Wa_22019338487

In reply to Greg Kroah-Hartman

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

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

From: Vinay Belgaumkar <[email protected]>

[ Upstream commit 84c0b4a00610afbde650fdb8ad6db0424f7b2cc3 ]

Limit GT max frequency to 2600MHz and wait for frequency to reduce before proceeding with a transient flush. This is really only needed for the transient flush: if L2 flush is needed due to 16023588340 then there's no need to do this additional wait since we are already using the bigger hammer.

v2: Use generic names, ensure user set max frequency requests wait for flush to complete (Rodrigo)
v3:
- User requests wait via wait_var_event_timeout (Lucas)
- Close races on flush + user requests (Lucas)
- Fix xe_guc_pc_remove_flush_freq_limit() being called on last gt rather than root gt (Lucas)
v4:
- Only apply the freq reducing part if a TDF is needed: L2 flush trumps the need for waiting a lower frequency

Fixes aaa08078e725 ("drm/xe/bmg: Apply Wa_22019338487")
(cherry picked from commit deea6a7d6d803d6bb874a3e6f1b312e560e6c6df)
---
 drivers/gpu/drm/xe/xe_device.c       |   8 +-
 drivers/gpu/drm/xe/xe_guc_pc.c       | 125 +++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_guc_pc.h       |   2 +
 drivers/gpu/drm/xe/xe_guc_pc_types.h |   2 +
 4 files changed, 135 insertions(+), 2 deletions(-)
drivers/gpu/drm/xe/xe_device.c +6 -2
index 38fdddd7262aa..f3123914b1abf 100644
@@ -38,6 +38,7 @@
 #include "xe_gt_printk.h"
 #include "xe_gt_sriov_vf.h"
 #include "xe_guc.h"
+#include "xe_guc_pc.h"
 #include "xe_hw_engine_group.h"
 #include "xe_hwmon.h"
 #include "xe_irq.h"
@@ -1057,11 +1058,14 @@ void xe_device_td_flush(struct xe_device *xe)
 		return;
 
 	root_gt = xe_root_mmio_gt(xe);
-	if (XE_WA(root_gt, 16023588340))
+	if (XE_WA(root_gt, 16023588340)) {
 		/* A transient flush is not sufficient: flush the L2 */
 		xe_device_l2_flush(xe);
-	else
+	} else {
+		xe_guc_pc_apply_flush_freq_limit(&root_gt->uc.guc.pc);
 		tdf_request_sync(xe);
+		xe_guc_pc_remove_flush_freq_limit(&root_gt->uc.guc.pc);
+	}
 }
 
 u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size)
drivers/gpu/drm/xe/xe_guc_pc.c +125 -0

159 lines not on this page · raw message

drivers/gpu/drm/xe/xe_guc_pc.h +2 -0
index 39102b79602fd..0302c7426ccde 100644
@@ -37,5 +37,7 @@ u64 xe_guc_pc_mc6_residency(struct xe_guc_pc *pc);
 void xe_guc_pc_init_early(struct xe_guc_pc *pc);
 int xe_guc_pc_restore_stashed_freq(struct xe_guc_pc *pc);
 void xe_guc_pc_raise_unslice(struct xe_guc_pc *pc);
+void xe_guc_pc_apply_flush_freq_limit(struct xe_guc_pc *pc);
+void xe_guc_pc_remove_flush_freq_limit(struct xe_guc_pc *pc);
 
 #endif /* _XE_GUC_PC_H_ */
drivers/gpu/drm/xe/xe_guc_pc_types.h +2 -0
index 2978ac9a249b5..c02053948a579 100644
@@ -15,6 +15,8 @@
 struct xe_guc_pc {
 	/** @bo: GGTT buffer object that is shared with GuC PC */
 	struct xe_bo *bo;
+	/** @flush_freq_limit: 1 when max freq changes are limited by driver */
+	atomic_t flush_freq_limit;
 	/** @rp0_freq: HW RP0 frequency - The Maximum one */
 	u32 rp0_freq;
 	/** @rpa_freq: HW RPa frequency - The Achievable one */
Signature · 1 line
2.39.5
author maintainer
#109 lore raw

PATCH 109/178 lib: test_objagg: Set error message in check_expect_hints_stats()

In reply to Greg Kroah-Hartman

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

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

From: Dan Carpenter <[email protected]>

[ Upstream commit e6ed134a4ef592fe1fd0cafac9683813b3c8f3e8 ]

Smatch complains that the error message isn't set in the caller:

lib/test_objagg.c:923 test_hints_case2()
error: uninitialized symbol 'errmsg'.

This static checker warning only showed up after a recent refactoring but the bug dates back to when the code was originally added. This likely doesn't affect anything in real life.

---
 lib/test_objagg.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
lib/test_objagg.c +3 -1
index d34df4306b874..222b39fc2629e 100644
@@ -899,8 +899,10 @@ static int check_expect_hints_stats(struct objagg_hints *objagg_hints,
 	int err;
 
 	stats = objagg_hints_stats_get(objagg_hints);
-	if (IS_ERR(stats))
+	if (IS_ERR(stats)) {
+		*errmsg = "objagg_hints_stats_get() failed.";
 		return PTR_ERR(stats);
+	}
 	err = __check_expect_stats(stats, expect_stats, errmsg);
 	objagg_stats_put(stats);
 	return err;
Signature · 1 line
2.39.5
author maintainer
#110 lore raw

PATCH 110/178 amd-xgbe: align CL37 AN sequence as per databook

In reply to Greg Kroah-Hartman

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

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

From: Raju Rangoju <[email protected]>

[ Upstream commit 42fd432fe6d320323215ebdf4de4d0d7e56e6792 ]

Update the Clause 37 Auto-Negotiation implementation to properly align with the PCS hardware specifications:
- Fix incorrect bit settings in Link Status and Link Duplex fields
- Implement missing sequence steps 2 and 7

These changes ensure CL37 auto-negotiation protocol follows the exact sequence patterns as specified in the hardware databook.

Fixes 1bf40ada6290 ("amd-xgbe: Add support for clause 37 auto-negotiation")
---
 drivers/net/ethernet/amd/xgbe/xgbe-common.h | 2 ++
 drivers/net/ethernet/amd/xgbe/xgbe-mdio.c   | 9 +++++++++
 drivers/net/ethernet/amd/xgbe/xgbe.h        | 4 ++--
 3 files changed, 13 insertions(+), 2 deletions(-)
drivers/net/ethernet/amd/xgbe/xgbe-common.h +2 -0
index 3b70f67376331..aa25a8a0a106f 100644
@@ -1373,6 +1373,8 @@
 #define MDIO_VEND2_CTRL1_SS13		BIT(13)
 #endif
 
+#define XGBE_VEND2_MAC_AUTO_SW		BIT(9)
+
 /* MDIO mask values */
 #define XGBE_AN_CL73_INT_CMPLT		BIT(0)
 #define XGBE_AN_CL73_INC_LINK		BIT(1)
drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +9 -0
index 07f4f3418d018..3316c719f9f8c 100644
@@ -375,6 +375,10 @@ static void xgbe_an37_set(struct xgbe_prv_data *pdata, bool enable,
 		reg |= MDIO_VEND2_CTRL1_AN_RESTART;
 
 	XMDIO_WRITE(pdata, MDIO_MMD_VEND2, MDIO_CTRL1, reg);
+
+	reg = XMDIO_READ(pdata, MDIO_MMD_VEND2, MDIO_PCS_DIG_CTRL);
+	reg |= XGBE_VEND2_MAC_AUTO_SW;
+	XMDIO_WRITE(pdata, MDIO_MMD_VEND2, MDIO_PCS_DIG_CTRL, reg);
 }
 
 static void xgbe_an37_restart(struct xgbe_prv_data *pdata)
@@ -1003,6 +1007,11 @@ static void xgbe_an37_init(struct xgbe_prv_data *pdata)
 
 	netif_dbg(pdata, link, pdata->netdev, "CL37 AN (%s) initialized\n",
 		  (pdata->an_mode == XGBE_AN_MODE_CL37) ? "BaseX" : "SGMII");
+
+	reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1);
+	reg &= ~MDIO_AN_CTRL1_ENABLE;
+	XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_CTRL1, reg);
+
 }
 
 static void xgbe_an73_init(struct xgbe_prv_data *pdata)
drivers/net/ethernet/amd/xgbe/xgbe.h +2 -2
index ed5d43c16d0e2..7526a0906b391 100644
@@ -292,12 +292,12 @@
 #define XGBE_LINK_TIMEOUT		5
 #define XGBE_KR_TRAINING_WAIT_ITER	50
 
-#define XGBE_SGMII_AN_LINK_STATUS	BIT(1)
+#define XGBE_SGMII_AN_LINK_DUPLEX	BIT(1)
 #define XGBE_SGMII_AN_LINK_SPEED	(BIT(2) | BIT(3))
 #define XGBE_SGMII_AN_LINK_SPEED_10	0x00
 #define XGBE_SGMII_AN_LINK_SPEED_100	0x04
 #define XGBE_SGMII_AN_LINK_SPEED_1000	0x08
-#define XGBE_SGMII_AN_LINK_DUPLEX	BIT(4)
+#define XGBE_SGMII_AN_LINK_STATUS	BIT(4)
 
 /* ECC correctable error notification window (seconds) */
 #define XGBE_ECC_LIMIT			60
Signature · 1 line
2.39.5
author maintainer
#111 lore raw

PATCH 111/178 enic: fix incorrect MTU comparison in enic_change_mtu()

In reply to Greg Kroah-Hartman

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

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

From: Alok Tiwari <[email protected]>

[ Upstream commit aaf2b2480375099c022a82023e1cd772bf1c6a5d ]

The comparison in enic_change_mtu() incorrectly used the current netdev->mtu instead of the new new_mtu value when warning about an MTU exceeding the port MTU. This could suppress valid warnings or issue incorrect ones.

Fix the condition and log to properly reflect the new_mtu.

---
 drivers/net/ethernet/cisco/enic/enic_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/cisco/enic/enic_main.c +2 -2
index c753c35b26ebd..5a8ca5be9ca00 100644
@@ -1864,10 +1864,10 @@ static int enic_change_mtu(struct net_device *netdev, int new_mtu)
 	if (enic_is_dynamic(enic) || enic_is_sriov_vf(enic))
 		return -EOPNOTSUPP;
 
-	if (netdev->mtu > enic->port_mtu)
+	if (new_mtu > enic->port_mtu)
 		netdev_warn(netdev,
 			    "interface MTU (%d) set higher than port MTU (%d)\n",
-			    netdev->mtu, enic->port_mtu);
+			    new_mtu, enic->port_mtu);
 
 	return _enic_change_mtu(netdev, new_mtu);
 }
Signature · 1 line
2.39.5
author maintainer
#102 lore raw

PATCH 102/178 smb: client: set missing retry flag in cifs_writev_callback()

In reply to Greg Kroah-Hartman

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

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

From: Paulo Alcantara <[email protected]>

[ Upstream commit 74ee76bea4b445c023d04806e0bcd78a912fd30b ]

Set NETFS_SREQ_NEED_RETRY flag to tell netfslib that the subreq needs to be retried.

---
 fs/smb/client/cifssmb.c | 1 +
 1 file changed, 1 insertion(+)
fs/smb/client/cifssmb.c +1 -0
index f9ccae5de5b88..0e509a0433fb6 100644
@@ -1715,6 +1715,7 @@ cifs_writev_callback(struct mid_q_entry *mid)
 		break;
 	case MID_REQUEST_SUBMITTED:
 	case MID_RETRY_NEEDED:
+		__set_bit(NETFS_SREQ_NEED_RETRY, &wdata->subreq.flags);
 		result = -EAGAIN;
 		break;
 	default:
Signature · 1 line
2.39.5
author maintainer
#136 lore raw

PATCH 136/178 wifi: mac80211: drop invalid source address OCB frames

In reply to Greg Kroah-Hartman

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

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

From: Johannes Berg <[email protected]>

[ Upstream commit d1b1a5eb27c4948e8811cf4dbb05aaf3eb10700c ]

In OCB, don't accept frames from invalid source addresses (and in particular don't try to create stations for them), drop the frames instead.

---
 net/mac80211/rx.c | 4 ++++
 1 file changed, 4 insertions(+)
net/mac80211/rx.c +4 -0
index 09beb65d6108b..e73431549ce77 100644
@@ -4432,6 +4432,10 @@ static bool ieee80211_accept_frame(struct ieee80211_rx_data *rx)
 		if (!multicast &&
 		    !ether_addr_equal(sdata->dev->dev_addr, hdr->addr1))
 			return false;
+		/* reject invalid/our STA address */
+		if (!is_valid_ether_addr(hdr->addr2) ||
+		    ether_addr_equal(sdata->dev->dev_addr, hdr->addr2))
+			return false;
 		if (!rx->sta) {
 			int rate_idx;
 			if (status->encoding != RX_ENC_LEGACY)
Signature · 1 line
2.39.5
author maintainer
#145 lore raw

PATCH 145/178 xhci: dbc: Flush queued requests before stopping dbc

In reply to Greg Kroah-Hartman

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

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

From: Mathias Nyman <[email protected]>

commit efe3e3ae5a66cb38ef29c909e951b4039044bae9 upstream.

Flush dbc requests when dbc is stopped and transfer rings are freed. Failure to flush them lead to leaking memory and dbc completing odd requests after resuming from suspend, leading to error messages such as:

[   95.344392] xhci_hcd 0000:00:0d.0: no matched request
Cc stable <[email protected]>
Fixes dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver")
---
 drivers/usb/host/xhci-dbgcap.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/usb/host/xhci-dbgcap.c +4 -0
@@ -652,6 +652,10 @@ static void xhci_dbc_stop(struct xhci_db
 	case DS_DISABLED:
 		return;
 	case DS_CONFIGURED:
+		spin_lock(&dbc->lock);
+		xhci_dbc_flush_requests(dbc);
+		spin_unlock(&dbc->lock);
+
 		if (dbc->driver->disconnect)
 			dbc->driver->disconnect(dbc);
 		break;
author maintainer
#146 lore raw

PATCH 146/178 xhci: Disable stream for xHC controller with XHCI_BROKEN_STREAMS

In reply to Greg Kroah-Hartman

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

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

From: Hongyu Xie <[email protected]>

commit cd65ee81240e8bc3c3119b46db7f60c80864b90b upstream.

Disable stream for platform xHC controller with broken stream.

Fixes 14aec589327a6 ("storage: accept some UAS devices if streams are unavailable")
Cc stable <[email protected]>
---
 drivers/usb/host/xhci-plat.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/usb/host/xhci-plat.c +2 -1
@@ -326,7 +326,8 @@ int xhci_plat_probe(struct platform_devi
 	}
 
 	usb3_hcd = xhci_get_usb3_hcd(xhci);
-	if (usb3_hcd && HCC_MAX_PSA(xhci->hcc_params) >= 4)
+	if (usb3_hcd && HCC_MAX_PSA(xhci->hcc_params) >= 4 &&
+	    !(xhci->quirks & XHCI_BROKEN_STREAMS))
 		usb3_hcd->can_do_streams = 1;
 
 	if (xhci->shared_hcd) {
author maintainer
#147 lore raw

PATCH 147/178 HID: appletb-kbd: fix memory corruption of input_handler_list

In reply to Greg Kroah-Hartman

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

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

From: Qasim Ijaz <[email protected]>

commit c80f2b047d5cc42fbd2dff9d1942d4ba7545100f upstream.

In appletb_kbd_probe an input handler is initialised and then registered with input core through input_register_handler(). When this happens input core will add the input handler (specifically its node) to the global input_handler_list. The input_handler_list is central to the functionality of input core and is traversed in various places in input core. An example of this is when a new input device is plugged in and gets registered with input core.

The input_handler in probe is allocated as device managed memory. If a probe failure occurs after input_register_handler() the input_handler memory is freed, yet it will remain in the input_handler_list. This effectively means the input_handler_list contains a dangling pointer to data belonging to a freed input handler.

This causes an issue when any other input device is plugged in - in my case I had an old PixArt HP USB optical mouse and I decided to plug it in after a failure occurred after input_register_handler(). This lead to the registration of this input device via input_register_device which involves traversing over every handler in the corrupted input_handler_list and calling input_attach_handler(), giving each handler a chance to bind to newly registered device.

The core of this bug is a UAF which causes memory corruption of input_handler_list and to fix it we must ensure the input handler is unregistered from input core, this is done through input_unregister_handler().

[   63.191597] ==================================================================
[   63.192094] BUG: KASAN: slab-use-after-free in input_attach_handler.isra.0+0x1a9/0x1e0
[   63.192094] Read of size 8 at addr ffff888105ea7c80 by task kworker/0:2/54
[   63.192094]
[   63.192094] CPU: 0 UID: 0 PID: 54 Comm: kworker/0:2 Not tainted 6.16.0-rc2-00321-g2aa6621d
[   63.192094] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-debian-1.164
[   63.192094] Workqueue: usb_hub_wq hub_event
[   63.192094] Call Trace:
[   63.192094]  <TASK>
[   63.192094]  dump_stack_lvl+0x53/0x70
[   63.192094]  print_report+0xce/0x670
[   63.192094]  kasan_report+0xce/0x100
[   63.192094]  input_attach_handler.isra.0+0x1a9/0x1e0
[   63.192094]  input_register_device+0x76c/0xd00
[   63.192094]  hidinput_connect+0x686d/0xad60
[   63.192094]  hid_connect+0xf20/0x1b10
[   63.192094]  hid_hw_start+0x83/0x100
[   63.192094]  hid_device_probe+0x2d1/0x680
[   63.192094]  really_probe+0x1c3/0x690
[   63.192094]  __driver_probe_device+0x247/0x300
[   63.192094]  driver_probe_device+0x49/0x210
[   63.192094]  __device_attach_driver+0x160/0x320
[   63.192094]  bus_for_each_drv+0x10f/0x190
[   63.192094]  __device_attach+0x18e/0x370
[   63.192094]  bus_probe_device+0x123/0x170
[   63.192094]  device_add+0xd4d/0x1460
[   63.192094]  hid_add_device+0x30b/0x910
[   63.192094]  usbhid_probe+0x920/0xe00
[   63.192094]  usb_probe_interface+0x363/0x9a0
[   63.192094]  really_probe+0x1c3/0x690
[   63.192094]  __driver_probe_device+0x247/0x300
[   63.192094]  driver_probe_device+0x49/0x210
[   63.192094]  __device_attach_driver+0x160/0x320
[   63.192094]  bus_for_each_drv+0x10f/0x190
[   63.192094]  __device_attach+0x18e/0x370
[   63.192094]  bus_probe_device+0x123/0x170
[   63.192094]  device_add+0xd4d/0x1460
[   63.192094]  usb_set_configuration+0xd14/0x1880
[   63.192094]  usb_generic_driver_probe+0x78/0xb0
[   63.192094]  usb_probe_device+0xaa/0x2e0
[   63.192094]  really_probe+0x1c3/0x690
[   63.192094]  __driver_probe_device+0x247/0x300
[   63.192094]  driver_probe_device+0x49/0x210
[   63.192094]  __device_attach_driver+0x160/0x320
[   63.192094]  bus_for_each_drv+0x10f/0x190
[   63.192094]  __device_attach+0x18e/0x370
[   63.192094]  bus_probe_device+0x123/0x170
[   63.192094]  device_add+0xd4d/0x1460
[   63.192094]  usb_new_device+0x7b4/0x1000
[   63.192094]  hub_event+0x234d/0x3fa0
[   63.192094]  process_one_work+0x5bf/0xfe0
[   63.192094]  worker_thread+0x777/0x13a0
[   63.192094]  </TASK>
[   63.192094]
[   63.192094] Allocated by task 54:
[   63.192094]  kasan_save_stack+0x33/0x60
[   63.192094]  kasan_save_track+0x14/0x30
[   63.192094]  __kasan_kmalloc+0x8f/0xa0
[   63.192094]  __kmalloc_node_track_caller_noprof+0x195/0x420
[   63.192094]  devm_kmalloc+0x74/0x1e0
[   63.192094]  appletb_kbd_probe+0x39/0x440
[   63.192094]  hid_device_probe+0x2d1/0x680
[   63.192094]  really_probe+0x1c3/0x690
[   63.192094]  __driver_probe_device+0x247/0x300
[   63.192094]  driver_probe_device+0x49/0x210
[   63.192094]  __device_attach_driver+0x160/0x320
[...]
[   63.192094]
[   63.192094] Freed by task 54:
[   63.192094]  kasan_save_stack+0x33/0x60
[   63.192094]  kasan_save_track+0x14/0x30
[   63.192094]  kasan_save_free_info+0x3b/0x60
[   63.192094]  __kasan_slab_free+0x37/0x50
[   63.192094]  kfree+0xcf/0x360
[   63.192094]  devres_release_group+0x1f8/0x3c0
[   63.192094]  hid_device_probe+0x315/0x680
[   63.192094]  really_probe+0x1c3/0x690
[   63.192094]  __driver_probe_device+0x247/0x300
[   63.192094]  driver_probe_device+0x49/0x210
[   63.192094]  __device_attach_driver+0x160/0x320

[...]

Fixes 7d62ba8deacf ("HID: hid-appletb-kbd: add support for fn toggle between media and function mode")
---
 drivers/hid/hid-appletb-kbd.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/hid/hid-appletb-kbd.c +3 -1
@@ -427,13 +427,15 @@ static int appletb_kbd_probe(struct hid_
 	ret = appletb_kbd_set_mode(kbd, appletb_tb_def_mode);
 	if (ret) {
 		dev_err_probe(dev, ret, "Failed to set touchbar mode\n");
-		goto close_hw;
+		goto unregister_handler;
 	}
 
 	hid_set_drvdata(hdev, kbd);
 
 	return 0;
 
+unregister_handler:
+	input_unregister_handler(&kbd->inp_handler);
 close_hw:
 	if (kbd->backlight_dev)
 		put_device(&kbd->backlight_dev->dev);
author maintainer
#148 lore raw

PATCH 148/178 Input: xpad - support Acer NGR 200 Controller

In reply to Greg Kroah-Hartman

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

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

From: Nilton Perim Neto <[email protected]>

commit 22c69d786ef8fb789c61ca75492a272774221324 upstream.

Add the NGR 200 Xbox 360 to the list of recognized controllers.

---
 drivers/input/joystick/xpad.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/input/joystick/xpad.c +2 -0
@@ -174,6 +174,7 @@ static const struct xpad_device {
 	{ 0x05fd, 0x107a, "InterAct 'PowerPad Pro' X-Box pad (Germany)", 0, XTYPE_XBOX },
 	{ 0x05fe, 0x3030, "Chic Controller", 0, XTYPE_XBOX },
 	{ 0x05fe, 0x3031, "Chic Controller", 0, XTYPE_XBOX },
+	{ 0x0502, 0x1305, "Acer NGR200", 0, XTYPE_XBOX },
 	{ 0x062a, 0x0020, "Logic3 Xbox GamePad", 0, XTYPE_XBOX },
 	{ 0x062a, 0x0033, "Competition Pro Steering Wheel", 0, XTYPE_XBOX },
 	{ 0x06a3, 0x0200, "Saitek Racing Wheel", 0, XTYPE_XBOX },
@@ -520,6 +521,7 @@ static const struct usb_device_id xpad_t
 	XPAD_XBOX360_VENDOR(0x045e),		/* Microsoft Xbox 360 controllers */
 	XPAD_XBOXONE_VENDOR(0x045e),		/* Microsoft Xbox One controllers */
 	XPAD_XBOX360_VENDOR(0x046d),		/* Logitech Xbox 360-style controllers */
+	XPAD_XBOX360_VENDOR(0x0502),		/* Acer Inc. Xbox 360 style controllers */
 	XPAD_XBOX360_VENDOR(0x056e),		/* Elecom JC-U3613M */
 	XPAD_XBOX360_VENDOR(0x06a3),		/* Saitek P3600 */
 	XPAD_XBOX360_VENDOR(0x0738),		/* Mad Catz Xbox 360 controllers */
author maintainer
#149 lore raw

PATCH 149/178 Input: iqs7222 - explicitly define number of external channels

In reply to Greg Kroah-Hartman

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

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

From: Jeff LaBundy <[email protected]>

commit 63f4970a1219b5256e8ea952096c86dab666d312 upstream.

The number of external channels is assumed to be a multiple of 10, but this is not the case for IQS7222D. As a result, some CRx pins are wrongly prevented from being assigned to some channels.

Address this problem by explicitly defining the number of external channels for cases in which the number of external channels is not equal to the total number of available channels.

---
 drivers/input/misc/iqs7222.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/input/misc/iqs7222.c +5 -2
@@ -301,6 +301,7 @@ struct iqs7222_dev_desc {
 	int allow_offset;
 	int event_offset;
 	int comms_offset;
+	int ext_chan;
 	bool legacy_gesture;
 	struct iqs7222_reg_grp_desc reg_grps[IQS7222_NUM_REG_GRPS];
 };
@@ -315,6 +316,7 @@ static const struct iqs7222_dev_desc iqs
 		.allow_offset = 9,
 		.event_offset = 10,
 		.comms_offset = 12,
+		.ext_chan = 10,
 		.reg_grps = {
 			[IQS7222_REG_GRP_STAT] = {
 				.base = IQS7222_SYS_STATUS,
@@ -373,6 +375,7 @@ static const struct iqs7222_dev_desc iqs
 		.allow_offset = 9,
 		.event_offset = 10,
 		.comms_offset = 12,
+		.ext_chan = 10,
 		.legacy_gesture = true,
 		.reg_grps = {
 			[IQS7222_REG_GRP_STAT] = {
@@ -2244,7 +2247,7 @@ static int iqs7222_parse_chan(struct iqs
 	const struct iqs7222_dev_desc *dev_desc = iqs7222->dev_desc;
 	struct i2c_client *client = iqs7222->client;
 	int num_chan = dev_desc->reg_grps[IQS7222_REG_GRP_CHAN].num_row;
-	int ext_chan = rounddown(num_chan, 10);
+	int ext_chan = dev_desc->ext_chan ? : num_chan;
 	int error, i;
 	u16 *chan_setup = iqs7222->chan_setup[chan_index];
 	u16 *sys_setup = iqs7222->sys_setup;
@@ -2445,7 +2448,7 @@ static int iqs7222_parse_sldr(struct iqs
 	const struct iqs7222_dev_desc *dev_desc = iqs7222->dev_desc;
 	struct i2c_client *client = iqs7222->client;
 	int num_chan = dev_desc->reg_grps[IQS7222_REG_GRP_CHAN].num_row;
-	int ext_chan = rounddown(num_chan, 10);
+	int ext_chan = dev_desc->ext_chan ? : num_chan;
 	int count, error, reg_offset, i;
 	u16 *event_mask = &iqs7222->sys_setup[dev_desc->event_offset];
 	u16 *sldr_setup = iqs7222->sldr_setup[sldr_index];
author maintainer
#150 lore raw

PATCH 150/178 usb: cdnsp: do not disable slot for disabled slot

In reply to Greg Kroah-Hartman

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

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

From: Peter Chen <[email protected]>

commit 7e2c421ef88e9da9c39e01496b7f5b0b354b42bc upstream.

It doesn't need to do it, and the related command event returns 'Slot Not Enabled Error' status.

Fixes 3d82904559f4 ("usb: cdnsp: cdns3 Add main part of Cadence USBSSP DRD Driver")
Cc stable <[email protected]>
---
 drivers/usb/cdns3/cdnsp-ring.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/usb/cdns3/cdnsp-ring.c +3 -1
@@ -772,7 +772,9 @@ static int cdnsp_update_port_id(struct c
 	}
 
 	if (port_id != old_port) {
-		cdnsp_disable_slot(pdev);
+		if (pdev->slot_id)
+			cdnsp_disable_slot(pdev);
+
 		pdev->active_port = port;
 		cdnsp_enable_slot(pdev);
 	}
author maintainer
#151 lore raw

PATCH 151/178 usb: cdnsp: Fix issue with CV Bad Descriptor test

In reply to Greg Kroah-Hartman

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

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

From: Pawel Laszczak <[email protected]>

commit 2831a81077f5162f104ba5a97a7d886eb371c21c upstream.

The SSP2 controller has extra endpoint state preserve bit (ESP) which setting causes that endpoint state will be preserved during Halt Endpoint command. It is used only for EP0.
Without this bit the Command Verifier "TD 9.10 Bad Descriptor Test" failed.
Setting this bit doesn't have any impact for SSP controller.

---
 drivers/usb/cdns3/cdnsp-debug.h  |    5 +++--
 drivers/usb/cdns3/cdnsp-ep0.c    |   18 +++++++++++++++---
 drivers/usb/cdns3/cdnsp-gadget.h |    6 ++++++
 drivers/usb/cdns3/cdnsp-ring.c   |    3 ++-
 4 files changed, 26 insertions(+), 6 deletions(-)
drivers/usb/cdns3/cdnsp-debug.h +3 -2
@@ -327,12 +327,13 @@ static inline const char *cdnsp_decode_t
 	case TRB_RESET_EP:
 	case TRB_HALT_ENDPOINT:
 		ret = scnprintf(str, size,
-				"%s: ep%d%s(%d) ctx %08x%08x slot %ld flags %c",
+				"%s: ep%d%s(%d) ctx %08x%08x slot %ld flags %c %c",
 				cdnsp_trb_type_string(type),
 				ep_num, ep_id % 2 ? "out" : "in",
 				TRB_TO_EP_INDEX(field3), field1, field0,
 				TRB_TO_SLOT_ID(field3),
-				field3 & TRB_CYCLE ? 'C' : 'c');
+				field3 & TRB_CYCLE ? 'C' : 'c',
+				field3 & TRB_ESP ? 'P' : 'p');
 		break;
 	case TRB_STOP_RING:
 		ret = scnprintf(str, size,
drivers/usb/cdns3/cdnsp-ep0.c +15 -3
@@ -414,6 +414,7 @@ static int cdnsp_ep0_std_request(struct
 void cdnsp_setup_analyze(struct cdnsp_device *pdev)
 {
 	struct usb_ctrlrequest *ctrl = &pdev->setup;
+	struct cdnsp_ep *pep;
 	int ret = -EINVAL;
 	u16 len;
 
@@ -427,10 +428,21 @@ void cdnsp_setup_analyze(struct cdnsp_de
 		goto out;
 	}
 
+	pep = &pdev->eps[0];
+
 	/* Restore the ep0 to Stopped/Running state. */
-	if (pdev->eps[0].ep_state & EP_HALTED) {
-		trace_cdnsp_ep0_halted("Restore to normal state");
-		cdnsp_halt_endpoint(pdev, &pdev->eps[0], 0);
+	if (pep->ep_state & EP_HALTED) {
+		if (GET_EP_CTX_STATE(pep->out_ctx) == EP_STATE_HALTED)
+			cdnsp_halt_endpoint(pdev, pep, 0);
+
+		/*
+		 * Halt Endpoint Command for SSP2 for ep0 preserve current
+		 * endpoint state and driver has to synchronize the
+		 * software endpoint state with endpoint output context
+		 * state.
+		 */
+		pep->ep_state &= ~EP_HALTED;
+		pep->ep_state |= EP_STOPPED;
 	}
 
 	/*
drivers/usb/cdns3/cdnsp-gadget.h +6 -0
@@ -987,6 +987,12 @@ enum cdnsp_setup_dev {
 #define STREAM_ID_FOR_TRB(p)		((((p)) << 16) & GENMASK(31, 16))
 #define SCT_FOR_TRB(p)			(((p) << 1) & 0x7)
 
+/*
+ * Halt Endpoint Command TRB field.
+ * The ESP bit only exists in the SSP2 controller.
+ */
+#define TRB_ESP				BIT(9)
+
 /* Link TRB specific fields. */
 #define TRB_TC				BIT(1)
 
drivers/usb/cdns3/cdnsp-ring.c +2 -1
@@ -2485,7 +2485,8 @@ void cdnsp_queue_halt_endpoint(struct cd
 {
 	cdnsp_queue_command(pdev, 0, 0, 0, TRB_TYPE(TRB_HALT_ENDPOINT) |
 			    SLOT_ID_FOR_TRB(pdev->slot_id) |
-			    EP_ID_FOR_TRB(ep_index));
+			    EP_ID_FOR_TRB(ep_index) |
+			    (!ep_index ? TRB_ESP : 0));
 }
 
 void cdnsp_force_header_wakeup(struct cdnsp_device *pdev, int intf_num)