linux-2.6-block.git
3 weeks agoMerge tag 'pci-v6.16-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci
Linus Torvalds [Sat, 28 Jun 2025 03:17:48 +0000 (20:17 -0700)]
Merge tag 'pci-v6.16-fixes-2' of git://git./linux/kernel/git/pci/pci

Pull PCI fix from Bjorn Helgaas:

 - Fix a PTM debugfs build error with CONFIG_DEBUG_FS=n &&
   CONFIG_PCIE_PTM=y (Manivannan Sadhasivam)

* tag 'pci-v6.16-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
  PCI/PTM: Build debugfs code only if CONFIG_DEBUG_FS is enabled

3 weeks agoMerge tag 'drm-fixes-2025-06-28' of https://gitlab.freedesktop.org/drm/kernel
Linus Torvalds [Sat, 28 Jun 2025 02:38:36 +0000 (19:38 -0700)]
Merge tag 'drm-fixes-2025-06-28' of https://gitlab.freedesktop.org/drm/kernel

Pull drm fixes from Dave Airlie:
 "Regular weekly drm updates, nothing out of the ordinary, amdgpu, xe,
  i915 and a few misc bits. Seems about right for this time in the
  release cycle.

  core:
   - fix drm_writeback_connector_cleanup function signature
   - use correct HDMI audio bridge in drm_connector_hdmi_audio_init

  bridge:
   - SN65DSI86: fix HPD

  amdgpu:
   - Cleaner shader support for additional GFX9 GPUs
   - MES firmware compatibility fixes
   - Discovery error reporting fixes
   - SDMA6/7 userq fixes
   - Backlight fix
   - EDID sanity check

  i915:
   - Fix for SNPS PHY HDMI for 1080p@120Hz
   - Correct DP AUX DPCD probe address
   - Followup build fix for GCOV and AutoFDO enabled config

  xe:
   - Missing error check
   - Fix xe_hwmon_power_max_write
   - Move flushes
   - Explicitly exit CT safe mode on unwind
   - Process deferred GGTT node removals on device unwind"

* tag 'drm-fixes-2025-06-28' of https://gitlab.freedesktop.org/drm/kernel:
  drm/xe: Process deferred GGTT node removals on device unwind
  drm/xe/guc: Explicitly exit CT safe mode on unwind
  drm/xe: move DPT l2 flush to a more sensible place
  drm/xe: Move DSB l2 flush to a more sensible place
  drm/bridge: ti-sn65dsi86: Add HPD for DisplayPort connector type
  drm/i915: fix build error some more
  drm/xe/hwmon: Fix xe_hwmon_power_max_write
  drm/xe/display: Add check for alloc_ordered_workqueue()
  drm/amd/display: Add sanity checks for drm_edid_raw()
  drm/amd/display: Fix AMDGPU_MAX_BL_LEVEL value
  drm/amdgpu/sdma7: add ucode version checks for userq support
  drm/amdgpu/sdma6: add ucode version checks for userq support
  drm/amd: Adjust output for discovery error handling
  drm/amdgpu/mes: add compatibility checks for set_hw_resource_1
  drm/amdgpu/gfx9: Add Cleaner Shader Support for GFX9.x GPUs
  drm/bridge-connector: Fix bridge in drm_connector_hdmi_audio_init()
  drm/dp: Change AUX DPCD probe address from DPCD_REV to LANE0_1_STATUS
  drm/i915/snps_hdmi_pll: Fix 64-bit divisor truncation by using div64_u64
  drm: writeback: Fix drm_writeback_connector_cleanup signature

3 weeks agoMerge tag 'cxl-fixes-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl
Linus Torvalds [Sat, 28 Jun 2025 00:58:32 +0000 (17:58 -0700)]
Merge tag 'cxl-fixes-6.16-rc4' of git://git./linux/kernel/git/cxl/cxl

Pull Compute Express Link (CXL) fixes from Dave Jiang:
 "These fixes address a few issues in the CXL subsystem, including
  dealing with some bugs in the CXL EDAC and RAS drivers:

   - Fix return value of cxlctl_validate_set_features()

   - Fix min_scrub_cycle of a region miscaculation and add additional
     documentation

   - Fix potential memory leak issues for CXL EDAC

   - Fix CPER handler device confusion for CXL RAS

   - Fix using wrong repair type to check DRAM event record"

* tag 'cxl-fixes-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl:
  cxl/edac: Fix using wrong repair type to check dram event record
  cxl/ras: Fix CPER handler device confusion
  cxl/edac: Fix potential memory leak issues
  cxl/Documentation: Add more description about min/max scrub cycle
  cxl/edac: Fix the min_scrub_cycle of a region miscalculation
  cxl: fix return value in cxlctl_validate_set_features()

3 weeks agoMerge tag 'libcrypto-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 28 Jun 2025 00:32:30 +0000 (17:32 -0700)]
Merge tag 'libcrypto-for-linus' of git://git./linux/kernel/git/ebiggers/linux

Pull crypto library fix from Eric Biggers:
 "Fix a regression where the purgatory code sometimes fails to build"

* tag 'libcrypto-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux:
  lib/crypto: sha256: Mark sha256_choose_blocks as __always_inline

3 weeks agoMerge tag 'drm-misc-fixes-2025-06-26' of https://gitlab.freedesktop.org/drm/misc...
Dave Airlie [Fri, 27 Jun 2025 20:53:00 +0000 (06:53 +1000)]
Merge tag 'drm-misc-fixes-2025-06-26' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes

drm-misc-fixes for v6.16-rc4:
- Fix function signature of drm_writeback_connector_cleanup.
- Use correct HDMI audio bridge in drm_connector_hdmi_audio_init.
- Make HPD work on SN65DSI86.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://lore.kernel.org/r/3dd1d5e1-73b6-4b0c-a208-f7d6235cf530@linux.intel.com
3 weeks agoMerge tag 'acpi-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Fri, 27 Jun 2025 19:08:36 +0000 (12:08 -0700)]
Merge tag 'acpi-6.16-rc4' of git://git./linux/kernel/git/rafael/linux-pm

Pull ACPI fix from Rafael Wysocki:
 "Revert a commit that attempted to fix a memory leak in an error code
  path and introduced a different issue (Zhe Qiao)"

* tag 'acpi-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  Revert "PCI/ACPI: Fix allocated memory release on error in pci_acpi_scan_root()"

3 weeks agoMerge tag 'block-6.16-20250626' of git://git.kernel.dk/linux
Linus Torvalds [Fri, 27 Jun 2025 16:02:33 +0000 (09:02 -0700)]
Merge tag 'block-6.16-20250626' of git://git.kernel.dk/linux

Pull block fixes from Jens Axboe:

 - Fixes for ublk:
      - fix C++ narrowing warnings in the uapi header
      - update/improve UBLK_F_SUPPORT_ZERO_COPY comment in uapi header
      - fix for the ublk ->queue_rqs() implementation, limiting a batch
        to just the specific task AND ring
      - ublk_get_data() error handling fix
      - sanity check more arguments in ublk_ctrl_add_dev()
      - selftest addition

 - NVMe pull request via Christoph:
      - reset delayed remove_work after reconnect
      - fix atomic write size validation

 - Fix for a warning introduced in bdev_count_inflight_rw() in this
   merge window

* tag 'block-6.16-20250626' of git://git.kernel.dk/linux:
  block: fix false warning in bdev_count_inflight_rw()
  ublk: sanity check add_dev input for underflow
  nvme: fix atomic write size validation
  nvme: refactor the atomic write unit detection
  nvme: reset delayed remove_work after reconnect
  ublk: setup ublk_io correctly in case of ublk_get_data() failure
  ublk: update UBLK_F_SUPPORT_ZERO_COPY comment in UAPI header
  ublk: fix narrowing warnings in UAPI header
  selftests: ublk: don't take same backing file for more than one ublk devices
  ublk: build batch from IOs in same io_ring_ctx and io task

3 weeks agoMerge tag 'io_uring-6.16-20250626' of git://git.kernel.dk/linux
Linus Torvalds [Fri, 27 Jun 2025 15:55:57 +0000 (08:55 -0700)]
Merge tag 'io_uring-6.16-20250626' of git://git.kernel.dk/linux

Pull io_uring fixes from Jens Axboe:

 - Two tweaks for a recent fix: fixing a memory leak if multiple iovecs
   were initially mapped but only the first was used and hence turned
   into a UBUF rathan than an IOVEC iterator, and catching a case where
   a retry would be done even if the previous segment wasn't full

 - Small series fixing an issue making the vm unhappy if debugging is
   turned on, hitting a VM_BUG_ON_PAGE()

 - Fix a resource leak in io_import_dmabuf() in the error handling case,
   which is a regression in this merge window

 - Mark fallocate as needing to be write serialized, as is already done
   for truncate and buffered writes

* tag 'io_uring-6.16-20250626' of git://git.kernel.dk/linux:
  io_uring/kbuf: flag partial buffer mappings
  io_uring/net: mark iov as dynamically allocated even for single segments
  io_uring: fix resource leak in io_import_dmabuf()
  io_uring: don't assume uaddr alignment in io_vec_fill_bvec
  io_uring/rsrc: don't rely on user vaddr alignment
  io_uring/rsrc: fix folio unpinning
  io_uring: make fallocate be hashed work

3 weeks agoMerge tag 'ata-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/libata...
Linus Torvalds [Fri, 27 Jun 2025 15:30:37 +0000 (08:30 -0700)]
Merge tag 'ata-6.16-rc4' of git://git./linux/kernel/git/libata/linux

Pull ata fix from Niklas Cassel:

 - Use the correct DMI identifier for ASUSPRO-D840SA LPM quirk such that
   the quirk actually gets applied (me)

* tag 'ata-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux:
  ata: ahci: Use correct DMI identifier for ASUSPRO-D840SA LPM quirk

3 weeks agoMerge tag 's390-6.16-3' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Linus Torvalds [Fri, 27 Jun 2025 15:26:25 +0000 (08:26 -0700)]
Merge tag 's390-6.16-3' of git://git./linux/kernel/git/s390/linux

Pull s390 fixes from Alexander Gordeev:

 - Fix incorrectly dropped dereferencing of the stack nth entry
   introduced with a previous KASAN false positive fix

 - Use a proper memdup_array_user() helper to prevent overflow in a
   protected key size calculation

* tag 's390-6.16-3' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390/ptrace: Fix pointer dereferencing in regs_get_kernel_stack_nth()
  s390/pkey: Prevent overflow in size calculation for memdup_user()

3 weeks agoMerge tag 'sound-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 27 Jun 2025 15:21:05 +0000 (08:21 -0700)]
Merge tag 'sound-6.16-rc4' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "A collection of small fixes again:

   - A regression fix for hibernation bug in ASoC SoundWire

   - Fixes for the new Qualcomm USB offload stuff

   - A potential OOB access fix in USB-audio

   - A potential memleadk fix in ASoC Intel

   - Quirks for HD-audio and ASoC AMD ACP"

* tag 'sound-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda/realtek: Fix built-in mic on ASUS VivoBook X507UAR
  ALSA: usb: qcom: fix NULL pointer dereference in qmi_stop_session
  ASoC: SOF: Intel: hda: Use devm_kstrdup() to avoid memleak.
  ASoC: rt721-sdca: fix boost gain calculation error
  ALSA: qc_audio_offload: Fix missing error code in prepare_qmi_response()
  ALSA: hda/realtek: Add mic-mute LED setup for ASUS UM5606
  ALSA: usb-audio: Fix out-of-bounds read in snd_usb_get_audioformat_uac3()
  ALSA: hda/realtek: fix mute/micmute LEDs for HP EliteBook 6 G1a
  ASoC: amd: ps: fix for soundwire failures during hibernation exit sequence
  ASoC: amd: yc: Add DMI quirk for Lenovo IdeaPad Slim 5 15
  ASoC: amd: yc: add quirk for Acer Nitro ANV15-41 internal mic
  ASoC: qcom: sm8250: Fix possibly undefined reference
  ALSA: hda/realtek - Enable mute LED on HP Pavilion Laptop 15-eg100
  ALSA: hda/realtek: Add quirks for some Clevo laptops

3 weeks agoMerge tag 'v6.16-p6' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Linus Torvalds [Fri, 27 Jun 2025 05:05:24 +0000 (22:05 -0700)]
Merge tag 'v6.16-p6' of git://git./linux/kernel/git/herbert/crypto-2.6

Pull crypto fix from Herbert Xu:
 "This fixes a regression where wp512 can no longer be used with hmac"

* tag 'v6.16-p6' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
  crypto: wp512 - Use API partial block handling

3 weeks agoMerge tag 'bcachefs-2025-06-26' of git://evilpiepirate.org/bcachefs
Linus Torvalds [Fri, 27 Jun 2025 02:49:12 +0000 (19:49 -0700)]
Merge tag 'bcachefs-2025-06-26' of git://evilpiepirate.org/bcachefs

Pull bcachefs fixes from Kent Overstreet:

 - Lots of small check/repair fixes, primarily in subvol loop and
   directory structure loop (when involving snapshots).

 - Fix a few 6.16 regressions: rare UAF in the foreground allocator path
   when taking a transaction restart from the transaction bump
   allocator, and some small fallout from the change to log the error
   being corrected in the journal when repairing errors, also some
   fallout from the btree node read error logging improvements.

   (Alan, Bharadwaj)

 - New option: journal_rewind

   This lets the entire filesystem be reset to an earlier point in time.

   Note that this is only a disaster recovery tool, and right now there
   are major caveats to using it (discards should be disabled, in
   particular), but it successfully restored the filesystem of one of
   the users who was bit by the subvolume deletion bug and didn't have
   backups. I'll likely be making some changes to the discard path in
   the future to make this a reliable recovery tool.

 - Some new btree iterator tracepoints, for tracking down some
   livelock-ish behaviour we've been seeing in the main data write path.

* tag 'bcachefs-2025-06-26' of git://evilpiepirate.org/bcachefs: (51 commits)
  bcachefs: Plumb correct ip to trans_relock_fail tracepoint
  bcachefs: Ensure we rewind to run recovery passes
  bcachefs: Ensure btree node scan runs before checking for scanned nodes
  bcachefs: btree_root_unreadable_and_scan_found_nothing should not be autofix
  bcachefs: fix bch2_journal_keys_peek_prev_min() underflow
  bcachefs: Use wait_on_allocator() when allocating journal
  bcachefs: Check for bad write buffer key when moving from journal
  bcachefs: Don't unlock the trans if ret doesn't match BCH_ERR_operation_blocked
  bcachefs: Fix range in bch2_lookup_indirect_extent() error path
  bcachefs: fix spurious error_throw
  bcachefs: Add missing bch2_err_class() to fileattr_set()
  bcachefs: Add missing key type checks to check_snapshot_exists()
  bcachefs: Don't log fsck err in the journal if doing repair elsewhere
  bcachefs: Fix *__bch2_trans_subbuf_alloc() error path
  bcachefs: Fix missing newlines before ero
  bcachefs: fix spurious error in read_btree_roots()
  bcachefs: fsck: Fix oops in key_visible_in_snapshot()
  bcachefs: fsck: fix unhandled restart in topology repair
  bcachefs: fsck: Fix check_directory_structure when no check_dirents
  bcachefs: Fix restart handling in btree_node_scrub_work()
  ...

3 weeks agoMerge tag 'hid-for-linus-2025062701' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 27 Jun 2025 00:06:01 +0000 (17:06 -0700)]
Merge tag 'hid-for-linus-2025062701' of git://git./linux/kernel/git/hid/hid

Pull HID fixes from Jiri Kosina:

 - fix for stalls during suspend/resume cycles with hid-nintendo (Daniel
   J. Ogorchock)

 - memory leak and reference count fixes in hid-wacom and in-appletb-kdb
   (Qasim Ijaz)

 - race condition (leading to kernel crash) fix during device removal in
   hid-wacom (Thomas Zeitlhofer)

 - fix for missed interrupt in intel-thc-hid (Intel-thc-hid:)

 - support for a bunch of new device IDs

* tag 'hid-for-linus-2025062701' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:
  HID: lenovo: Add support for ThinkPad X1 Tablet Thin Keyboard Gen2
  HID: appletb-kbd: fix "appletb_backlight" backlight device reference counting
  HID: wacom: fix crash in wacom_aes_battery_handler()
  HID: intel-ish-hid: ipc: Add Wildcat Lake PCI device ID
  hid: intel-ish-hid: Use PCI_DEVICE_DATA() macro for ISH device table
  HID: lenovo: Restrict F7/9/11 mode to compact keyboards only
  HID: Add IGNORE quirk for SMARTLINKTECHNOLOGY
  HID: input: lower message severity of 'No inputs registered, leaving' to debug
  HID: quirks: Add quirk for 2 Chicony Electronics HP 5MP Cameras
  HID: Intel-thc-hid: Intel-quicki2c: Enhance QuickI2C reset flow
  HID: nintendo: avoid bluetooth suspend/resume stalls
  HID: wacom: fix kobject reference count leak
  HID: wacom: fix memory leak on sysfs attribute creation failure
  HID: wacom: fix memory leak on kobject creation failure

3 weeks agoMerge tag 'drm-xe-fixes-2025-06-26' of https://gitlab.freedesktop.org/drm/xe/kernel...
Dave Airlie [Thu, 26 Jun 2025 23:13:45 +0000 (09:13 +1000)]
Merge tag 'drm-xe-fixes-2025-06-26' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes

UAPI Changes:
Driver Changes:
- Missing error check (Haoxiang Li)
- Fix xe_hwmon_power_max_write (Karthik)
- Move flushes (Maarten and Matthew Auld)
- Explicitly exit CT safe mode on unwind (Michal)
- Process deferred GGTT node removals on device unwind (Michal)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Hellstrom <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/aF1T6EzzC3xj4K4H@fedora
3 weeks agoMerge tag 'drm-intel-fixes-2025-06-26' of https://gitlab.freedesktop.org/drm/i915...
Dave Airlie [Thu, 26 Jun 2025 23:09:01 +0000 (09:09 +1000)]
Merge tag 'drm-intel-fixes-2025-06-26' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-fixes

- Fix for SNPS PHY HDMI for 1080p@120Hz
- Correct DP AUX DPCD probe address
- Followup build fix for GCOV and AutoFDO enabled config

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://lore.kernel.org/r/aFzsHR9WLYsxg8jy@jlahtine-mobl
3 weeks agoMerge tag 'amd-drm-fixes-6.16-2025-25-25' of https://gitlab.freedesktop.org/agd5f...
Dave Airlie [Thu, 26 Jun 2025 23:03:42 +0000 (09:03 +1000)]
Merge tag 'amd-drm-fixes-6.16-2025-25-25' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.16-2025-25-25:

amdgpu:
- Cleaner shader support for additional GFX9 GPUs
- MES firmware compatibility fixes
- Discovery error reporting fixes
- SDMA6/7 userq fixes
- Backlight fix
- EDID sanity check

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250625151734.11537-1-alexander.deucher@amd.com
3 weeks agoMerge tag 'devicetree-fixes-for-6.16-1' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 26 Jun 2025 19:26:39 +0000 (12:26 -0700)]
Merge tag 'devicetree-fixes-for-6.16-1' of git://git./linux/kernel/git/robh/linux

Pull devicetree fixes from Rob Herring:

 - Convert altr,uart-1.0 and altr,juart-1.0 to DT schema. These were
   applied for nios2, but never sent upstream.

 - Fix extra '/' in fsl,ls1028a-reset '$id' path

 - Fix warnings in ti,sn65dsi83 schema due to unnecessary $ref.

* tag 'devicetree-fixes-for-6.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
  dt-bindings: serial: Convert altr,uart-1.0 to DT schema
  dt-bindings: serial: Convert altr,juart-1.0 to DT schema
  dt-bindings: soc: fsl,ls1028a-reset: Drop extra "/" in $id
  dt-bindings: drm/bridge: ti-sn65dsi83: drop $ref to fix lvds-vod* warnings

3 weeks agoio_uring/kbuf: flag partial buffer mappings io_uring-6.16-20250626
Jens Axboe [Thu, 26 Jun 2025 18:17:48 +0000 (12:17 -0600)]
io_uring/kbuf: flag partial buffer mappings

A previous commit aborted mapping more for a non-incremental ring for
bundle peeking, but depending on where in the process this peeking
happened, it would not necessarily prevent a retry by the user. That can
create gaps in the received/read data.

Add struct buf_sel_arg->partial_map, which can pass this information
back. The networking side can then map that to internal state and use it
to gate retry as well.

Since this necessitates a new flag, change io_sr_msg->retry to a
retry_flags member, and store both the retry and partial map condition
in there.

Cc: stable@vger.kernel.org
Fixes: 26ec15e4b0c1 ("io_uring/kbuf: don't truncate end buffer for multiple buffer peeks")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoMerge tag 'net-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Thu, 26 Jun 2025 16:13:27 +0000 (09:13 -0700)]
Merge tag 'net-6.16-rc4' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Paolo Abeni:
 "Including fixes from bluetooth and wireless.

  Current release - regressions:

   - bridge: fix use-after-free during router port configuration

  Current release - new code bugs:

   - eth: wangxun: fix the creation of page_pool

  Previous releases - regressions:

   - netpoll: initialize UDP checksum field before checksumming

   - wifi: mac80211: finish link init before RCU publish

   - bluetooth: fix use-after-free in vhci_flush()

   - eth:
      - ionic: fix DMA mapping test
      - bnxt: properly flush XDP redirect lists

  Previous releases - always broken:

   - netlink: specs: enforce strict naming of properties

   - unix: don't leave consecutive consumed OOB skbs.

   - vsock: fix linux/vm_sockets.h userspace compilation errors

   - selftests: fix TCP packet checksum"

* tag 'net-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (38 commits)
  net: libwx: fix the creation of page_pool
  net: selftests: fix TCP packet checksum
  atm: Release atm_dev_mutex after removing procfs in atm_dev_deregister().
  netlink: specs: enforce strict naming of properties
  netlink: specs: tc: replace underscores with dashes in names
  netlink: specs: rt-link: replace underscores with dashes in names
  netlink: specs: mptcp: replace underscores with dashes in names
  netlink: specs: ovs_flow: replace underscores with dashes in names
  netlink: specs: devlink: replace underscores with dashes in names
  netlink: specs: dpll: replace underscores with dashes in names
  netlink: specs: ethtool: replace underscores with dashes in names
  netlink: specs: fou: replace underscores with dashes in names
  netlink: specs: nfsd: replace underscores with dashes in names
  net: enetc: Correct endianness handling in _enetc_rd_reg64
  atm: idt77252: Add missing `dma_map_error()`
  bnxt: properly flush XDP redirect lists
  vsock/uapi: fix linux/vm_sockets.h userspace compilation errors
  wifi: mac80211: finish link init before RCU publish
  wifi: iwlwifi: mvm: assume '1' as the default mac_config_cmd version
  selftest: af_unix: Add tests for -ECONNRESET.
  ...

3 weeks agoblock: fix false warning in bdev_count_inflight_rw() block-6.16-20250626
Yu Kuai [Thu, 26 Jun 2025 11:57:43 +0000 (19:57 +0800)]
block: fix false warning in bdev_count_inflight_rw()

While bdev_count_inflight is interating all cpus, if some IOs are issued
from traversed cpu and then completed from the cpu that is not traversed
yet:

cpu0
cpu1
bdev_count_inflight
 //for_each_possible_cpu
 // cpu0 is 0
 infliht += 0
// issue a io
blk_account_io_start
// cpu0 inflight ++

cpu2
// the io is done
blk_account_io_done
// cpu2 inflight --
 // cpu 1 is 0
 inflight += 0
 // cpu2 is -1
 inflight += -1
 ...

In this case, the total inflight will be -1, causing lots of false
warning. Fix the problem by removing the warning.

Noted there is still a valid warning for nvme-mpath(From Yi) that is not
fixed yet.

Fixes: f5482ee5edb9 ("block: WARN if bdev inflight counter is negative")
Reported-by: Yi Zhang <yi.zhang@redhat.com>
Closes: https://lore.kernel.org/linux-block/aFtUXy-lct0WxY2w@mozart.vkv.me/T/#mae89155a5006463d0a21a4a2c35ae0034b26a339
Reported-and-tested-by: Calvin Owens <calvin@wbinvd.org>
Closes: https://lore.kernel.org/linux-block/aFtUXy-lct0WxY2w@mozart.vkv.me/T/#m1d935a00070bf95055d0ac84e6075158b08acaef
Reported-by: Dave Chinner <david@fromorbit.com>
Closes: https://lore.kernel.org/linux-block/aFuypjqCXo9-5_En@dread.disaster.area/
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20250626115743.1641443-1-yukuai3@huawei.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoMerge tag 'nvme-6.16-2025-06-26' of git://git.infradead.org/nvme into block-6.16
Jens Axboe [Thu, 26 Jun 2025 13:31:52 +0000 (07:31 -0600)]
Merge tag 'nvme-6.16-2025-06-26' of git://git.infradead.org/nvme into block-6.16

Pull NVMe fixes from Christoph:

" - reset delayed remove_work after reconnect (Keith Busch)
  - fix atomic write size validation (Christoph Hellwig)"

* tag 'nvme-6.16-2025-06-26' of git://git.infradead.org/nvme:
  nvme: fix atomic write size validation
  nvme: refactor the atomic write unit detection
  nvme: reset delayed remove_work after reconnect

3 weeks agoublk: sanity check add_dev input for underflow
Ronnie Sahlberg [Thu, 26 Jun 2025 02:20:45 +0000 (12:20 +1000)]
ublk: sanity check add_dev input for underflow

Add additional checks that queue depth and number of queues are
non-zero.

Signed-off-by: Ronnie Sahlberg <rsahlberg@whamcloud.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250626022046.235018-1-ronniesahlberg@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agodrm/xe: Process deferred GGTT node removals on device unwind
Michal Wajdeczko [Thu, 12 Jun 2025 22:09:36 +0000 (00:09 +0200)]
drm/xe: Process deferred GGTT node removals on device unwind

While we are indirectly draining our dedicated workqueue ggtt->wq
that we use to complete asynchronous removal of some GGTT nodes,
this happends as part of the managed-drm unwinding (ggtt_fini_early),
which could be later then manage-device unwinding, where we could
already unmap our MMIO/GMS mapping (mmio_fini).

This was recently observed during unsuccessful VF initialization:

 [ ] xe 0000:00:02.1: probe with driver xe failed with error -62
 [ ] xe 0000:00:02.1: DEVRES REL ffff88811e747340 __xe_bo_unpin_map_no_vm (16 bytes)
 [ ] xe 0000:00:02.1: DEVRES REL ffff88811e747540 __xe_bo_unpin_map_no_vm (16 bytes)
 [ ] xe 0000:00:02.1: DEVRES REL ffff88811e747240 __xe_bo_unpin_map_no_vm (16 bytes)
 [ ] xe 0000:00:02.1: DEVRES REL ffff88811e747040 tiles_fini (16 bytes)
 [ ] xe 0000:00:02.1: DEVRES REL ffff88811e746840 mmio_fini (16 bytes)
 [ ] xe 0000:00:02.1: DEVRES REL ffff88811e747f40 xe_bo_pinned_fini (16 bytes)
 [ ] xe 0000:00:02.1: DEVRES REL ffff88811e746b40 devm_drm_dev_init_release (16 bytes)
 [ ] xe 0000:00:02.1: [drm:drm_managed_release] drmres release begin
 [ ] xe 0000:00:02.1: [drm:drm_managed_release] REL ffff88810ef81640 __fini_relay (8 bytes)
 [ ] xe 0000:00:02.1: [drm:drm_managed_release] REL ffff88810ef80d40 guc_ct_fini (8 bytes)
 [ ] xe 0000:00:02.1: [drm:drm_managed_release] REL ffff88810ef80040 __drmm_mutex_release (8 bytes)
 [ ] xe 0000:00:02.1: [drm:drm_managed_release] REL ffff88810ef80140 ggtt_fini_early (8 bytes)

and this was leading to:

 [ ] BUG: unable to handle page fault for address: ffffc900058162a0
 [ ] #PF: supervisor write access in kernel mode
 [ ] #PF: error_code(0x0002) - not-present page
 [ ] Oops: Oops: 0002 [#1] SMP NOPTI
 [ ] Tainted: [W]=WARN
 [ ] Workqueue: xe-ggtt-wq ggtt_node_remove_work_func [xe]
 [ ] RIP: 0010:xe_ggtt_set_pte+0x6d/0x350 [xe]
 [ ] Call Trace:
 [ ]  <TASK>
 [ ]  xe_ggtt_clear+0xb0/0x270 [xe]
 [ ]  ggtt_node_remove+0xbb/0x120 [xe]
 [ ]  ggtt_node_remove_work_func+0x30/0x50 [xe]
 [ ]  process_one_work+0x22b/0x6f0
 [ ]  worker_thread+0x1e8/0x3d

Add managed-device action that will explicitly drain the workqueue
with all pending node removals prior to releasing MMIO/GSM mapping.

Fixes: 919bb54e989c ("drm/xe: Fix missing runtime outer protection for ggtt_remove_node")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250612220937.857-2-michal.wajdeczko@intel.com
(cherry picked from commit 89d2835c3680ab1938e22ad81b1c9f8c686bd391)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
3 weeks agodrm/xe/guc: Explicitly exit CT safe mode on unwind
Michal Wajdeczko [Thu, 12 Jun 2025 22:09:37 +0000 (00:09 +0200)]
drm/xe/guc: Explicitly exit CT safe mode on unwind

During driver probe we might be briefly using CT safe mode, which
is based on a delayed work, but usually we are able to stop this
once we have IRQ fully operational.  However, if we abort the probe
quite early then during unwind we might try to destroy the workqueue
while there is still a pending delayed work that attempts to restart
itself which triggers a WARN.

This was recently observed during unsuccessful VF initialization:

 [ ] xe 0000:00:02.1: probe with driver xe failed with error -62
 [ ] ------------[ cut here ]------------
 [ ] workqueue: cannot queue safe_mode_worker_func [xe] on wq xe-g2h-wq
 [ ] WARNING: CPU: 9 PID: 0 at kernel/workqueue.c:2257 __queue_work+0x287/0x710
 [ ] RIP: 0010:__queue_work+0x287/0x710
 [ ] Call Trace:
 [ ]  delayed_work_timer_fn+0x19/0x30
 [ ]  call_timer_fn+0xa1/0x2a0

Exit the CT safe mode on unwind to avoid that warning.

Fixes: 09b286950f29 ("drm/xe/guc: Allow CTB G2H processing without G2H IRQ")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250612220937.857-3-michal.wajdeczko@intel.com
(cherry picked from commit 2ddbb73ec20b98e70a5200cb85deade22ccea2ec)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
3 weeks agodrm/xe: move DPT l2 flush to a more sensible place
Matthew Auld [Fri, 6 Jun 2025 10:45:48 +0000 (11:45 +0100)]
drm/xe: move DPT l2 flush to a more sensible place

Only need the flush for DPT host updates here. Normal GGTT updates don't
need special flush.

Fixes: 01570b446939 ("drm/xe/bmg: implement Wa_16023588340")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: stable@vger.kernel.org # v6.12+
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250606104546.1996818-4-matthew.auld@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit 35db1da40c8cfd7511dc42f342a133601eb45449)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
3 weeks agodrm/xe: Move DSB l2 flush to a more sensible place
Maarten Lankhorst [Fri, 6 Jun 2025 10:45:47 +0000 (11:45 +0100)]
drm/xe: Move DSB l2 flush to a more sensible place

Flushing l2 is only needed after all data has been written.

Fixes: 01570b446939 ("drm/xe/bmg: implement Wa_16023588340")
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: stable@vger.kernel.org # v6.12+
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250606104546.1996818-3-matthew.auld@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit 0dd2dd0182bc444a62652e89d08c7f0e4fde15ba)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
3 weeks agonvme: fix atomic write size validation
Christoph Hellwig [Wed, 11 Jun 2025 04:54:56 +0000 (06:54 +0200)]
nvme: fix atomic write size validation

Don't mix the namespace and controller values, and validate the
per-controller limit when probing the controller.  This avoid spurious
failures for controllers with namespaces that have different namespaces
with different logical block sizes, or report the per-namespace values
only for some namespaces.

It also fixes a missing queue_limits_cancel_update in an error path by
removing that error path.

Fixes: 8695f060a029 ("nvme: all namespaces in a subsystem must adhere to a common atomic write size")
Reported-by: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: John Garry <john.g.garry@oracle.com>
Tested-by: Yi Zhang <yi.zhang@redhat.com>
3 weeks agonvme: refactor the atomic write unit detection
Christoph Hellwig [Wed, 11 Jun 2025 05:09:21 +0000 (07:09 +0200)]
nvme: refactor the atomic write unit detection

Move all the code out of nvme_update_disk_info into the helper, and
rename the helper to have a somewhat less clumsy name.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: John Garry <john.g.garry@oracle.com>
3 weeks agonvme: reset delayed remove_work after reconnect
Keith Busch [Wed, 11 Jun 2025 04:50:47 +0000 (06:50 +0200)]
nvme: reset delayed remove_work after reconnect

The remove_work will proceed with permanently disconnecting on the
initial final path failure if the head shows no paths after the delay.
If a new path connects while the remove_work is pending, and if that new
path happens to disconnect before that remove_work executes, the delayed
removal should reset based on the most recent path disconnect time, but
queue_delayed_work() won't do anything if the work is already pending.
Attempt to cancel the delayed work when a new path connects, and use
mod_delayed_work() in case the remove_work remains pending anyway.

Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
3 weeks agonet: libwx: fix the creation of page_pool
Jiawen Wu [Wed, 25 Jun 2025 02:39:24 +0000 (10:39 +0800)]
net: libwx: fix the creation of page_pool

'rx_ring->size' means the count of ring descriptors multiplied by the
size of one descriptor. When increasing the count of ring descriptors,
it may exceed the limit of pool size.

[ 864.209610] page_pool_create_percpu() gave up with errno -7
[ 864.209613] txgbe 0000:11:00.0: Page pool creation failed: -7

Fix to set the pool_size to the count of ring descriptors.

Fixes: 850b971110b2 ("net: libwx: Allocate Rx and Tx resources")
Cc: stable@vger.kernel.org
Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/434C72BFB40E350A+20250625023924.21821-1-jiawenwu@trustnetic.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agonet: selftests: fix TCP packet checksum
Jakub Kicinski [Tue, 24 Jun 2025 18:32:58 +0000 (11:32 -0700)]
net: selftests: fix TCP packet checksum

The length in the pseudo header should be the length of the L3 payload
AKA the L4 header+payload. The selftest code builds the packet from
the lower layers up, so all the headers are pushed already when it
constructs L4. We need to subtract the lower layer headers from skb->len.

Fixes: 3e1e58d64c3d ("net: add generic selftest support")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reported-by: Oleksij Rempel <o.rempel@pengutronix.de>
Tested-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://patch.msgid.link/20250624183258.3377740-1-kuba@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agoALSA: hda/realtek: Fix built-in mic on ASUS VivoBook X507UAR
Salvatore Bonaccorso [Wed, 25 Jun 2025 18:41:28 +0000 (20:41 +0200)]
ALSA: hda/realtek: Fix built-in mic on ASUS VivoBook X507UAR

The built-in mic of ASUS VivoBook X507UAR is broken recently by the fix
of the pin sort. The fixup ALC256_FIXUP_ASUS_MIC_NO_PRESENCE is working
for addressing the regression, too.

Fixes: 3b4309546b48 ("ALSA: hda: Fix headset detection failure due to unstable sort")
Reported-by: Igor Tamara <igor.tamara@gmail.com>
Closes: https://bugs.debian.org/1108069
Signed-off-by: Salvatore Bonaccorso <carnil@debian.org>
Link: https://lore.kernel.org/CADdHDco7_o=4h_epjEAb92Dj-vUz_PoTC2-W9g5ncT2E0NzfeQ@mail.gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
3 weeks agoMerge tag 'asoc-fix-v6.16-rc3' of https://git.kernel.org/pub/scm/linux/kernel/git...
Takashi Iwai [Thu, 26 Jun 2025 05:59:33 +0000 (07:59 +0200)]
Merge tag 'asoc-fix-v6.16-rc3' of https://git./linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v6.16

A small collection of fixes, the main one being a fix for resume from
hibernation on AMD systems, plus a few new quirk entries for AMD
systems.

3 weeks agoMerge tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Linus Torvalds [Thu, 26 Jun 2025 04:09:02 +0000 (21:09 -0700)]
Merge tag 'bpf-fixes' of git://git./linux/kernel/git/bpf/bpf

Pull bpf fixes from Alexei Starovoitov:

 - Fix use-after-free in libbpf when map is resized (Adin Scannell)

 - Fix verifier assumptions about 2nd argument of bpf_sysctl_get_name
   (Jerome Marchand)

 - Fix verifier assumption of nullness of d_inode in dentry (Song Liu)

 - Fix global starvation of LRU map (Willem de Bruijn)

 - Fix potential NULL dereference in btf_dump__free (Yuan Chen)

* tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
  selftests/bpf: adapt one more case in test_lru_map to the new target_free
  libbpf: Fix possible use-after-free for externs
  selftests/bpf: Convert test_sysctl to prog_tests
  bpf: Specify access type of bpf_sysctl_get_name args
  libbpf: Fix null pointer dereference in btf_dump__free on allocation failure
  bpf: Adjust free target to avoid global starvation of LRU map
  bpf: Mark dentry->d_inode as trusted_or_null

3 weeks agobcachefs: Plumb correct ip to trans_relock_fail tracepoint
Kent Overstreet [Wed, 25 Jun 2025 18:55:59 +0000 (14:55 -0400)]
bcachefs: Plumb correct ip to trans_relock_fail tracepoint

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
3 weeks agobcachefs: Ensure we rewind to run recovery passes
Kent Overstreet [Wed, 25 Jun 2025 04:48:14 +0000 (00:48 -0400)]
bcachefs: Ensure we rewind to run recovery passes

Fix a 6.16 regression from the recovery pass rework, which introduced a
bug where calling bch2_run_explicit_recovery_pass() would only return
the error code to rewind recovery for the first call that scheduled that
recovery pass.

If the error code from the first call was swallowed (because it was
called by an asynchronous codepath), subsequent calls would go "ok, this
pass is already marked as needing to run" and return 0.

Fixing this ensures that check_topology bails out to run btree_node_scan
before doing any repair.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
3 weeks agobcachefs: Ensure btree node scan runs before checking for scanned nodes
Kent Overstreet [Wed, 25 Jun 2025 16:45:11 +0000 (12:45 -0400)]
bcachefs: Ensure btree node scan runs before checking for scanned nodes

Previously, calling bch2_btree_has_scanned_nodes() when btree node
scan hadn't actually run would erroniously return false - causing us to
think a btree was entirely gone.

This fixes a 6.16 regression from moving the scheduling of btree node
scan out of bch2_btree_lost_data() (fixing the bug where we'd schedule
it persistently in the superblock) and only scheduling it when
check_toploogy() is asking for scanned btree nodes.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
3 weeks agobcachefs: btree_root_unreadable_and_scan_found_nothing should not be autofix
Kent Overstreet [Wed, 25 Jun 2025 16:17:51 +0000 (12:17 -0400)]
bcachefs: btree_root_unreadable_and_scan_found_nothing should not be autofix

Autofix is specified in btree_gc.c if it's not an important btree.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
3 weeks agoMerge tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Thu, 26 Jun 2025 03:48:48 +0000 (20:48 -0700)]
Merge tag 'pull-fixes' of git://git./linux/kernel/git/viro/vfs

Pull mount fixes from Al Viro:
 "Several mount-related fixes"

* tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  userns and mnt_idmap leak in open_tree_attr(2)
  attach_recursive_mnt(): do not lock the covering tree when sliding something under it
  replace collect_mounts()/drop_collected_mounts() with a safer variant

3 weeks agoatm: Release atm_dev_mutex after removing procfs in atm_dev_deregister().
Kuniyuki Iwashima [Tue, 24 Jun 2025 21:45:00 +0000 (14:45 -0700)]
atm: Release atm_dev_mutex after removing procfs in atm_dev_deregister().

syzbot reported a warning below during atm_dev_register(). [0]

Before creating a new device and procfs/sysfs for it, atm_dev_register()
looks up a duplicated device by __atm_dev_lookup().  These operations are
done under atm_dev_mutex.

However, when removing a device in atm_dev_deregister(), it releases the
mutex just after removing the device from the list that __atm_dev_lookup()
iterates over.

So, there will be a small race window where the device does not exist on
the device list but procfs/sysfs are still not removed, triggering the
splat.

Let's hold the mutex until procfs/sysfs are removed in
atm_dev_deregister().

[0]:
proc_dir_entry 'atm/atmtcp:0' already registered
WARNING: CPU: 0 PID: 5919 at fs/proc/generic.c:377 proc_register+0x455/0x5f0 fs/proc/generic.c:377
Modules linked in:
CPU: 0 UID: 0 PID: 5919 Comm: syz-executor284 Not tainted 6.16.0-rc2-syzkaller-00047-g52da431bf03b #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
RIP: 0010:proc_register+0x455/0x5f0 fs/proc/generic.c:377
Code: 48 89 f9 48 c1 e9 03 80 3c 01 00 0f 85 a2 01 00 00 48 8b 44 24 10 48 c7 c7 20 c0 c2 8b 48 8b b0 d8 00 00 00 e8 0c 02 1c ff 90 <0f> 0b 90 90 48 c7 c7 80 f2 82 8e e8 0b de 23 09 48 8b 4c 24 28 48
RSP: 0018:ffffc9000466fa30 EFLAGS: 00010282
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffffff817ae248
RDX: ffff888026280000 RSI: ffffffff817ae255 RDI: 0000000000000001
RBP: ffff8880232bed48 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000001 R12: ffff888076ed2140
R13: dffffc0000000000 R14: ffff888078a61340 R15: ffffed100edda444
FS:  00007f38b3b0c6c0(0000) GS:ffff888124753000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f38b3bdf953 CR3: 0000000076d58000 CR4: 00000000003526f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 proc_create_data+0xbe/0x110 fs/proc/generic.c:585
 atm_proc_dev_register+0x112/0x1e0 net/atm/proc.c:361
 atm_dev_register+0x46d/0x890 net/atm/resources.c:113
 atmtcp_create+0x77/0x210 drivers/atm/atmtcp.c:369
 atmtcp_attach drivers/atm/atmtcp.c:403 [inline]
 atmtcp_ioctl+0x2f9/0xd60 drivers/atm/atmtcp.c:464
 do_vcc_ioctl+0x12c/0x930 net/atm/ioctl.c:159
 sock_do_ioctl+0x115/0x280 net/socket.c:1190
 sock_ioctl+0x227/0x6b0 net/socket.c:1311
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:907 [inline]
 __se_sys_ioctl fs/ioctl.c:893 [inline]
 __x64_sys_ioctl+0x18b/0x210 fs/ioctl.c:893
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xcd/0x4c0 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f38b3b74459
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 51 18 00 00 90 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 b0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f38b3b0c198 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00007f38b3bfe318 RCX: 00007f38b3b74459
RDX: 0000000000000000 RSI: 0000000000006180 RDI: 0000000000000005
RBP: 00007f38b3bfe310 R08: 65732f636f72702f R09: 65732f636f72702f
R10: 65732f636f72702f R11: 0000000000000246 R12: 00007f38b3bcb0ac
R13: 00007f38b3b0c1a0 R14: 0000200000000200 R15: 00007f38b3bcb03b
 </TASK>

Fixes: 64bf69ddff76 ("[ATM]: deregistration removes device from atm_devs list immediately")
Reported-by: syzbot+8bd335d2ad3b93e80715@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/685316de.050a0220.216029.0087.GAE@google.com/
Tested-by: syzbot+8bd335d2ad3b93e80715@syzkaller.appspotmail.com
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20250624214505.570679-1-kuni1840@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agoMerge branch 'netlink-specs-enforce-strict-naming-of-properties'
Jakub Kicinski [Wed, 25 Jun 2025 22:36:30 +0000 (15:36 -0700)]
Merge branch 'netlink-specs-enforce-strict-naming-of-properties'

Jakub Kicinski says:

====================
netlink: specs: enforce strict naming of properties

I got annoyed once again by the name properties in the ethtool spec
which use underscore instead of dash. I previously assumed that there
is a lot of such properties in the specs so fixing them now would
be near impossible. On a closer look, however, I only found 22
(rough grep suggests we have ~4.8k names in the specs, so bad ones
are just 0.46%).

Add a regex to the JSON schema to enforce the naming, fix the few
bad names. I was hoping we could start enforcing this from newer
families, but there's no correlation between the protocol and the
number of errors. If anything classic netlink has more recently
added specs so it has fewer errors.

The regex is just for name properties which will end up visible
to the user (in Python or YNL CLI). I left the c-name properties
alone, those don't matter as much. C codegen rewrites them, anyway.

I'm not updating the spec for genetlink-c. Looks like it has no
users, new families use genetlink, all old ones need genetlink-legacy.
If these patches are merged I will remove genetlink-c completely
in net-next.
====================

Link: https://patch.msgid.link/20250624211002.3475021-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: enforce strict naming of properties
Jakub Kicinski [Tue, 24 Jun 2025 21:10:02 +0000 (14:10 -0700)]
netlink: specs: enforce strict naming of properties

Add a regexp to make sure all names which may end up being visible
to the user consist of lower case characters, numbers and dashes.
Underscores keep sneaking into the specs, which is not visible
in the C code but makes the Python and alike inconsistent.

Note that starting with a number is okay, as in C the full
name will include the family name.

For legacy families we can't enforce the naming in the family
name or the multicast group names, as these are part of the
binary uAPI of the kernel.

For classic netlink we need to allow capital letters in names
of struct members. TC has some structs with capitalized members.

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250624211002.3475021-11-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: tc: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:10:01 +0000 (14:10 -0700)]
netlink: specs: tc: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: a1bcfde83669 ("doc/netlink/specs: Add a spec for tc")
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250624211002.3475021-10-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: rt-link: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:10:00 +0000 (14:10 -0700)]
netlink: specs: rt-link: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: b2f63d904e72 ("doc/netlink: Add spec for rt link messages")
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250624211002.3475021-9-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: mptcp: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:09:59 +0000 (14:09 -0700)]
netlink: specs: mptcp: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: bc8aeb2045e2 ("Documentation: netlink: add a YAML spec for mptcp")
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250624211002.3475021-8-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: ovs_flow: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:09:58 +0000 (14:09 -0700)]
netlink: specs: ovs_flow: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: 93b230b549bc ("netlink: specs: add ynl spec for ovs_flow")
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Eelco Chaudron <echaudro@redhat.com>
Link: https://patch.msgid.link/20250624211002.3475021-7-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: devlink: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:09:57 +0000 (14:09 -0700)]
netlink: specs: devlink: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: 429ac6211494 ("devlink: define enum for attr types of dynamic attributes")
Fixes: f2f9dd164db0 ("netlink: specs: devlink: add the remaining command to generate complete split_ops")
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250624211002.3475021-6-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: dpll: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:09:56 +0000 (14:09 -0700)]
netlink: specs: dpll: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: 3badff3a25d8 ("dpll: spec: Add Netlink spec in YAML")
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250624211002.3475021-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: ethtool: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:09:55 +0000 (14:09 -0700)]
netlink: specs: ethtool: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen replaces special chars in names)
but gives more uniform naming in Python.

Fixes: 13e59344fb9d ("net: ethtool: add support for symmetric-xor RSS hash")
Fixes: 46fb3ba95b93 ("ethtool: Add an interface for flashing transceiver modules' firmware")
Reviewed-by: Kory Maincent <kory.maincent@bootlin.com>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250624211002.3475021-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: fou: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:09:54 +0000 (14:09 -0700)]
netlink: specs: fou: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: 4eb77b4ecd3c ("netlink: add a proto specification for FOU")
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250624211002.3475021-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonetlink: specs: nfsd: replace underscores with dashes in names
Jakub Kicinski [Tue, 24 Jun 2025 21:09:53 +0000 (14:09 -0700)]
netlink: specs: nfsd: replace underscores with dashes in names

We're trying to add a strict regexp for the name format in the spec.
Underscores will not be allowed, dashes should be used instead.
This makes no difference to C (codegen, if used, replaces special
chars in names) but it gives more uniform naming in Python.

Fixes: 13727f85b49b ("NFSD: introduce netlink stubs")
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Link: https://patch.msgid.link/20250624211002.3475021-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonet: enetc: Correct endianness handling in _enetc_rd_reg64
Simon Horman [Tue, 24 Jun 2025 16:35:12 +0000 (17:35 +0100)]
net: enetc: Correct endianness handling in _enetc_rd_reg64

enetc_hw.h provides two versions of _enetc_rd_reg64.
One which simply calls ioread64() when available.
And another that composes the 64-bit result from ioread32() calls.

In the second case the code appears to assume that each ioread32() call
returns a little-endian value. However both the shift and logical or
used to compose the return value would not work correctly on big endian
systems if this were the case. Moreover, this is inconsistent with the
first case where the return value of ioread64() is assumed to be in host
byte order.

It appears that the correct approach is for both versions to treat the
return value of ioread*() functions as being in host byte order. And
this patch corrects the ioread32()-based version to do so.

This is a bug but would only manifest on big endian systems
that make use of the ioread32-based implementation of _enetc_rd_reg64.
While all in-tree users of this driver are little endian and
make use of the ioread64-based implementation of _enetc_rd_reg64.
Thus, no in-tree user of this driver is affected by this bug.

Flagged by Sparse.
Compile tested only.

Fixes: 16eb4c85c964 ("enetc: Add ethtool statistics")
Closes: https://lore.kernel.org/all/AM9PR04MB850500D3FC24FE23DEFCEA158879A@AM9PR04MB8505.eurprd04.prod.outlook.com/
Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20250624-etnetc-le-v1-1-a73a95d96e4e@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agoatm: idt77252: Add missing `dma_map_error()`
Thomas Fourier [Tue, 24 Jun 2025 06:41:47 +0000 (08:41 +0200)]
atm: idt77252: Add missing `dma_map_error()`

The DMA map functions can fail and should be tested for errors.

Signed-off-by: Thomas Fourier <fourier.thomas@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250624064148.12815-3-fourier.thomas@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agoselftests/bpf: adapt one more case in test_lru_map to the new target_free
Willem de Bruijn [Wed, 25 Jun 2025 21:03:55 +0000 (17:03 -0400)]
selftests/bpf: adapt one more case in test_lru_map to the new target_free

The below commit that updated BPF_MAP_TYPE_LRU_HASH free target,
also updated tools/testing/selftests/bpf/test_lru_map to match.

But that missed one case that passes with 4 cores, but fails at
higher cpu counts.

Update test_lru_sanity3 to also adjust its expectation of target_free.

This time tested with 1, 4, 16, 64 and 384 cpu count.

Fixes: d4adf1c9ee77 ("bpf: Adjust free target to avoid global starvation of LRU map")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20250625210412.2732970-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
3 weeks agolibbpf: Fix possible use-after-free for externs
Adin Scannell [Wed, 25 Jun 2025 05:02:15 +0000 (22:02 -0700)]
libbpf: Fix possible use-after-free for externs

The `name` field in `obj->externs` points into the BTF data at initial
open time. However, some functions may invalidate this after opening and
before loading (e.g. `bpf_map__set_value_size`), which results in
pointers into freed memory and undefined behavior.

The simplest solution is to simply `strdup` these strings, similar to
the `essent_name`, and free them at the same time.

In order to test this path, the `global_map_resize` BPF selftest is
modified slightly to ensure the presence of an extern, which causes this
test to fail prior to the fix. Given there isn't an obvious API or error
to test against, I opted to add this to the existing test as an aspect
of the resizing feature rather than duplicate the test.

Fixes: 9d0a23313b1a ("libbpf: Add capability for resizing datasec maps")
Signed-off-by: Adin Scannell <amscanne@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20250625050215.2777374-1-amscanne@meta.com
3 weeks agocxl/edac: Fix using wrong repair type to check dram event record
Li Ming [Fri, 20 Jun 2025 05:29:24 +0000 (13:29 +0800)]
cxl/edac: Fix using wrong repair type to check dram event record

cxl_find_rec_dram() is used to find a DRAM event record based on the
inputted attributes. Different repair_type of the inputted attributes
will check the DRAM event record in different ways.
When EDAC driver is performing a memory rank sparing, it should use
CXL_RANK_SPARING rather than CXL_BANK_SPARING as repair_type for DRAM
event record checking.

Fixes: 588ca944c277 ("cxl/edac: Add CXL memory device memory sparing control feature")
Signed-off-by: Li Ming <ming.li@zohomail.com>
Reviewed-by: Shiju Jose <shiju.jose@huawei.com>
Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Link: https://patch.msgid.link/20250620052924.138892-1-ming.li@zohomail.com
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
3 weeks agoMerge tag 'spi-fix-v6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/brooni...
Linus Torvalds [Wed, 25 Jun 2025 18:54:04 +0000 (11:54 -0700)]
Merge tag 'spi-fix-v6.16-rc3' of git://git./linux/kernel/git/broonie/spi

Pull spi fix from Mark Brown:
 "One fix for a runtime PM underflow when removing the Cadence QuadSPI
  driver"

* tag 'spi-fix-v6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
  spi: spi-cadence-quadspi: Fix pm runtime unbalance

3 weeks agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Wed, 25 Jun 2025 18:20:14 +0000 (11:20 -0700)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "Fixes all in drivers.

  ufs and megaraid_sas are small and obvious.

  The large diffstat in fnic comes from two pieces: the addition of
  quite a bit of logging (no change to function) and the reworking of
  the timeout allocation path for the two conditions that can occur
  simultaneously to prevent reusing the same abort frame and then both
  trying to free it"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: fnic: Fix missing DMA mapping error in fnic_send_frame()
  scsi: fnic: Set appropriate logging level for log message
  scsi: fnic: Add and improve logs in FDMI and FDMI ABTS paths
  scsi: fnic: Turn off FDMI ACTIVE flags on link down
  scsi: fnic: Fix crash in fnic_wq_cmpl_handler when FDMI times out
  scsi: ufs: core: Fix clk scaling to be conditional in reset and restore
  scsi: megaraid_sas: Fix invalid node index

3 weeks agoMerge tag 'uml-for-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/uml...
Linus Torvalds [Wed, 25 Jun 2025 18:13:31 +0000 (11:13 -0700)]
Merge tag 'uml-for-6.16-rc4' of git://git./linux/kernel/git/uml/linux

Pull UML fixes from Johannes Berg:

 - fix FP registers in seccomp mode

 - prevent duplicate devices in VFIO support

 - don't ignore errors in UBD thread start

 - reduce stack use with clang 19

* tag 'uml-for-6.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/uml/linux:
  um: vector: Reduce stack usage in vector_eth_configure()
  um: Use correct data source in fpregs_legacy_set()
  um: vfio: Prevent duplicate device assignments
  um: ubd: Add missing error check in start_io_thread()

3 weeks agoMerge tag 'wireless-2025-06-25' of https://git.kernel.org/pub/scm/linux/kernel/git...
Jakub Kicinski [Wed, 25 Jun 2025 17:26:16 +0000 (10:26 -0700)]
Merge tag 'wireless-2025-06-25' of https://git./linux/kernel/git/wireless/wireless

Johannes Berg says:

====================
Just a few fixes:
 - iwlegacy: work around large stack with clang/kasan
 - mac80211: fix integer overflow
 - mac80211: fix link struct init vs. RCU publish
 - iwlwifi: fix warning on IFF_UP

* tag 'wireless-2025-06-25' of https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless:
  wifi: mac80211: finish link init before RCU publish
  wifi: iwlwifi: mvm: assume '1' as the default mac_config_cmd version
  wifi: mac80211: fix beacon interval calculation overflow
  wifi: iwlegacy: work around excessive stack usage on clang/kasan
====================

Link: https://patch.msgid.link/20250625115433.41381-3-johannes@sipsolutions.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agoio_uring/net: mark iov as dynamically allocated even for single segments
Jens Axboe [Wed, 25 Jun 2025 16:17:06 +0000 (10:17 -0600)]
io_uring/net: mark iov as dynamically allocated even for single segments

A bigger array of vecs could've been allocated, but
io_ring_buffers_peek() still decided to cap the mapped range depending
on how much data was available. Hence don't rely on the segment count
to know if the request should be marked as needing cleanup, always
check upfront if the iov array is different than the fast_iov array.

Fixes: 26ec15e4b0c1 ("io_uring/kbuf: don't truncate end buffer for multiple buffer peeks")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agodrm/bridge: ti-sn65dsi86: Add HPD for DisplayPort connector type
Jayesh Choudhary [Tue, 24 Jun 2025 04:48:35 +0000 (10:18 +0530)]
drm/bridge: ti-sn65dsi86: Add HPD for DisplayPort connector type

By default, HPD was disabled on SN65DSI86 bridge. When the driver was
added (commit "a095f15c00e27"), the HPD_DISABLE bit was set in pre-enable
call which was moved to other function calls subsequently.
Later on, commit "c312b0df3b13" added detect utility for DP mode. But with
HPD_DISABLE bit set, all the HPD events are disabled[0] and the debounced
state always return 1 (always connected state).

Set HPD_DISABLE bit conditionally based on display sink's connector type.
Since the HPD_STATE is reflected correctly only after waiting for debounce
time (~100-400ms) and adding this delay in detect() is not feasible
owing to the performace impact (glitches and frame drop), remove runtime
calls in detect() and add hpd_enable()/disable() bridge hooks with runtime
calls, to detect hpd properly without any delay.

[0]: <https://www.ti.com/lit/gpn/SN65DSI86> (Pg. 32)

Fixes: c312b0df3b13 ("drm/bridge: ti-sn65dsi86: Implement bridge connector operations for DP")
Cc: Max Krummenacher <max.krummenacher@toradex.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Ernest Van Hoecke <ernest.vanhoecke@toradex.com>
Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250624044835.165708-1-j-choudhary@ti.com
3 weeks agoio_uring: fix resource leak in io_import_dmabuf()
Penglei Jiang [Wed, 25 Jun 2025 10:27:03 +0000 (03:27 -0700)]
io_uring: fix resource leak in io_import_dmabuf()

Replace the return statement with setting ret = -EINVAL and jumping to
the err label to ensure resources are released via io_release_dmabuf.

Fixes: a5c98e942457 ("io_uring/zcrx: dmabuf backed zerocopy receive")
Signed-off-by: Penglei Jiang <superman.xpt@gmail.com>
Link: https://lore.kernel.org/r/20250625102703.68336-1-superman.xpt@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoata: ahci: Use correct DMI identifier for ASUSPRO-D840SA LPM quirk
Niklas Cassel [Tue, 24 Jun 2025 07:40:30 +0000 (09:40 +0200)]
ata: ahci: Use correct DMI identifier for ASUSPRO-D840SA LPM quirk

ASUS store the board name in DMI_PRODUCT_NAME rather than
DMI_PRODUCT_VERSION. (Apparently it is only Lenovo that stores the
model-name in DMI_PRODUCT_VERSION.)

Use the correct DMI identifier, DMI_PRODUCT_NAME, to match the
ASUSPRO-D840SA board, such that the quirk actually gets applied.

Cc: stable@vger.kernel.org
Reported-by: Andy Yang <andyybtc79@gmail.com>
Tested-by: Andy Yang <andyybtc79@gmail.com>
Closes: https://lore.kernel.org/linux-ide/aFb3wXAwJSSJUB7o@ryzen/
Fixes: b5acc3628898 ("ata: ahci: Disallow LPM for ASUSPRO-D840SA motherboard")
Reviewed-by: Hans de Goede <hansg@kernel.org>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Link: https://lore.kernel.org/r/20250624074029.963028-2-cassel@kernel.org
Signed-off-by: Niklas Cassel <cassel@kernel.org>
3 weeks agoMerge tag 'iwlwifi-fixes-2025-06-25' of https://git.kernel.org/pub/scm/linux/kernel...
Johannes Berg [Wed, 25 Jun 2025 08:19:39 +0000 (10:19 +0200)]
Merge tag 'iwlwifi-fixes-2025-06-25' of https://git./linux/kernel/git/iwlwifi/iwlwifi-next

Miri Korenblit says:
====================
iwlwifi-fixes: fix failure in interface up
====================

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
3 weeks agoum: vector: Reduce stack usage in vector_eth_configure()
Tiwei Bie [Mon, 23 Jun 2025 11:08:29 +0000 (19:08 +0800)]
um: vector: Reduce stack usage in vector_eth_configure()

When compiling with clang (19.1.7), initializing *vp using a compound
literal may result in excessive stack usage. Fix it by initializing the
required fields of *vp individually.

Without this patch:

$ objdump -d arch/um/drivers/vector_kern.o | ./scripts/checkstack.pl x86_64 0
...
0x0000000000000540 vector_eth_configure [vector_kern.o]:1472
...

With this patch:

$ objdump -d arch/um/drivers/vector_kern.o | ./scripts/checkstack.pl x86_64 0
...
0x0000000000000540 vector_eth_configure [vector_kern.o]:208
...

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202506221017.WtB7Usua-lkp@intel.com/
Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com>
Link: https://patch.msgid.link/20250623110829.314864-1-tiwei.btw@antgroup.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
3 weeks agoum: Use correct data source in fpregs_legacy_set()
Tiwei Bie [Fri, 6 Jun 2025 12:44:28 +0000 (20:44 +0800)]
um: Use correct data source in fpregs_legacy_set()

Read from the buffer pointed to by 'from' instead of '&buf', as
'buf' contains no valid data when 'ubuf' is NULL.

Fixes: b1e1bd2e6943 ("um: Add helper functions to get/set state for SECCOMP")
Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com>
Link: https://patch.msgid.link/20250606124428.148164-5-tiwei.btw@antgroup.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
3 weeks agoum: vfio: Prevent duplicate device assignments
Tiwei Bie [Fri, 6 Jun 2025 12:44:27 +0000 (20:44 +0800)]
um: vfio: Prevent duplicate device assignments

Ensure devices are assigned only once. Reject subsequent requests
for duplicate assignments.

Fixes: a0e2cb6a9063 ("um: Add VFIO-based virtual PCI driver")
Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com>
Link: https://patch.msgid.link/20250606124428.148164-4-tiwei.btw@antgroup.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
3 weeks agoum: ubd: Add missing error check in start_io_thread()
Tiwei Bie [Fri, 6 Jun 2025 12:44:25 +0000 (20:44 +0800)]
um: ubd: Add missing error check in start_io_thread()

The subsequent call to os_set_fd_block() overwrites the previous
return value. OR the two return values together to fix it.

Fixes: f88f0bdfc32f ("um: UBD Improvements")
Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com>
Link: https://patch.msgid.link/20250606124428.148164-2-tiwei.btw@antgroup.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
3 weeks agodrm/i915: fix build error some more
Arnd Bergmann [Fri, 20 Jun 2025 11:18:18 +0000 (13:18 +0200)]
drm/i915: fix build error some more

An earlier patch fixed a build failure with clang, but I still see the
same problem with some configurations using gcc:

drivers/gpu/drm/i915/i915_pmu.c: In function 'config_mask':
include/linux/compiler_types.h:568:38: error: call to '__compiletime_assert_462' declared with attribute error: BUILD_BUG_ON failed: bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1
drivers/gpu/drm/i915/i915_pmu.c:116:3: note: in expansion of macro 'BUILD_BUG_ON'
  116 |   BUILD_BUG_ON(bit >

As I understand it, the problem is that the function is not always fully
inlined, but the __builtin_constant_p() can still evaluate the argument
as being constant.

Marking it as __always_inline so far works for me in all configurations.

Fixes: a7137b1825b5 ("drm/i915/pmu: Fix build error with GCOV and AutoFDO enabled")
Fixes: a644fde77ff7 ("drm/i915/pmu: Change bitmask of enabled events to u32")
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20250620111824.3395007-1-arnd@kernel.org
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
(cherry picked from commit ef69f9dd1cd7301cdf04ba326ed28152a3affcf6)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
3 weeks agoALSA: usb: qcom: fix NULL pointer dereference in qmi_stop_session
Pei Xiao [Tue, 24 Jun 2025 09:00:47 +0000 (17:00 +0800)]
ALSA: usb: qcom: fix NULL pointer dereference in qmi_stop_session

The find_substream() call may return NULL, but the error path
dereferenced 'subs' unconditionally via dev_err(&subs->dev->dev, ...),
causing a NULL pointer dereference when subs is NULL.

Fix by switching to &uadev[idx].udev->dev which is always valid
in this context.

Signed-off-by: Pei Xiao <xiaopei01@kylinos.cn>
Link: https://patch.msgid.link/86ac2939273ac853535049e60391c09d7688714e.1750755508.git.xiaopei01@kylinos.cn
Signed-off-by: Takashi Iwai <tiwai@suse.de>
3 weeks agoio_uring: don't assume uaddr alignment in io_vec_fill_bvec
Pavel Begunkov [Tue, 24 Jun 2025 13:40:35 +0000 (14:40 +0100)]
io_uring: don't assume uaddr alignment in io_vec_fill_bvec

There is no guaranteed alignment for user pointers. Don't use mask
trickery and adjust the offset by bv_offset.

Cc: stable@vger.kernel.org
Reported-by: David Hildenbrand <david@redhat.com>
Fixes: 9ef4cbbcb4ac3 ("io_uring: add infra for importing vectored reg buffers")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/io-uring/19530391f5c361a026ac9b401ff8e123bde55d98.1750771718.git.asml.silence@gmail.com/
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoio_uring/rsrc: don't rely on user vaddr alignment
Pavel Begunkov [Tue, 24 Jun 2025 13:40:34 +0000 (14:40 +0100)]
io_uring/rsrc: don't rely on user vaddr alignment

There is no guaranteed alignment for user pointers, however the
calculation of an offset of the first page into a folio after coalescing
uses some weird bit mask logic, get rid of it.

Cc: stable@vger.kernel.org
Reported-by: David Hildenbrand <david@redhat.com>
Fixes: a8edbb424b139 ("io_uring/rsrc: enable multi-hugepage buffer coalescing")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/io-uring/e387b4c78b33f231105a601d84eefd8301f57954.1750771718.git.asml.silence@gmail.com/
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoio_uring/rsrc: fix folio unpinning
Pavel Begunkov [Tue, 24 Jun 2025 13:40:33 +0000 (14:40 +0100)]
io_uring/rsrc: fix folio unpinning

syzbot complains about an unmapping failure:

[  108.070381][   T14] kernel BUG at mm/gup.c:71!
[  108.070502][   T14] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[  108.123672][   T14] Hardware name: QEMU KVM Virtual Machine, BIOS edk2-20250221-8.fc42 02/21/2025
[  108.127458][   T14] Workqueue: iou_exit io_ring_exit_work
[  108.174205][   T14] Call trace:
[  108.175649][   T14]  sanity_check_pinned_pages+0x7cc/0x7d0 (P)
[  108.178138][   T14]  unpin_user_page+0x80/0x10c
[  108.180189][   T14]  io_release_ubuf+0x84/0xf8
[  108.182196][   T14]  io_free_rsrc_node+0x250/0x57c
[  108.184345][   T14]  io_rsrc_data_free+0x148/0x298
[  108.186493][   T14]  io_sqe_buffers_unregister+0x84/0xa0
[  108.188991][   T14]  io_ring_ctx_free+0x48/0x480
[  108.191057][   T14]  io_ring_exit_work+0x764/0x7d8
[  108.193207][   T14]  process_one_work+0x7e8/0x155c
[  108.195431][   T14]  worker_thread+0x958/0xed8
[  108.197561][   T14]  kthread+0x5fc/0x75c
[  108.199362][   T14]  ret_from_fork+0x10/0x20

We can pin a tail page of a folio, but then io_uring will try to unpin
the head page of the folio. While it should be fine in terms of keeping
the page actually alive, mm folks say it's wrong and triggers a debug
warning. Use unpin_user_folio() instead of unpin_user_page*.

Cc: stable@vger.kernel.org
Debugged-by: David Hildenbrand <david@redhat.com>
Reported-by: syzbot+1d335893772467199ab6@syzkaller.appspotmail.com
Closes: https://lkml.kernel.org/r/683f1551.050a0220.55ceb.0017.GAE@google.com
Fixes: a8edbb424b139 ("io_uring/rsrc: enable multi-hugepage buffer coalescing")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/io-uring/a28b0f87339ac2acf14a645dad1e95bbcbf18acd.1750771718.git.asml.silence@gmail.com/
[axboe: adapt to current tree, massage commit message]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoublk: setup ublk_io correctly in case of ublk_get_data() failure
Ming Lei [Tue, 24 Jun 2025 10:41:21 +0000 (18:41 +0800)]
ublk: setup ublk_io correctly in case of ublk_get_data() failure

If ublk_get_data() fails, -EIOCBQUEUED is returned and the current command
becomes ASYNC. And the only reason is that mapping data can't move on,
because of no enough pages or pending signal, then the current ublk request
has to be requeued.

Once the request need to be requeued, we have to setup `ublk_io` correctly,
including io->cmd and flags, otherwise the request may not be forwarded to
ublk server successfully.

Fixes: 9810362a57cb ("ublk: don't call ublk_dispatch_req() for NEED_GET_DATA")
Reported-by: Changhui Zhong <czhong@redhat.com>
Closes: https://lore.kernel.org/linux-block/CAGVVp+VN9QcpHUz_0nasFf5q9i1gi8H8j-G-6mkBoqa3TyjRHA@mail.gmail.com/
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Changhui Zhong <czhong@redhat.com>
Link: https://lore.kernel.org/r/20250624104121.859519-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoublk: update UBLK_F_SUPPORT_ZERO_COPY comment in UAPI header
Caleb Sander Mateos [Sat, 21 Jun 2025 17:10:14 +0000 (11:10 -0600)]
ublk: update UBLK_F_SUPPORT_ZERO_COPY comment in UAPI header

UBLK_F_SUPPORT_ZERO_COPY has a very old comment describing the initial
idea for how zero-copy would be implemented. The actual implementation
added in commit 1f6540e2aabb ("ublk: zc register/unregister bvec") uses
io_uring registered buffers rather than shared memory mapping.
Remove the inaccurate remarks about mapping ublk request memory into the
ublk server's address space and requiring 4K block size. Replace them
with a description of the current zero-copy mechanism.

Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250621171015.354932-1-csander@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoublk: fix narrowing warnings in UAPI header
Caleb Sander Mateos [Sat, 21 Jun 2025 16:28:41 +0000 (10:28 -0600)]
ublk: fix narrowing warnings in UAPI header

When a C++ file compiled with -Wc++11-narrowing includes the UAPI header
linux/ublk_cmd.h, ublk_sqe_addr_to_auto_buf_reg()'s assignments of u64
values to u8, u16, and u32 fields result in compiler warnings. Add
explicit casts to the intended types to avoid these warnings. Drop the
unnecessary bitmasks.

Reported-by: Uday Shankar <ushankar@purestorage.com>
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Fixes: 99c1e4eb6a3f ("ublk: register buffer to local io_uring with provided buf index via UBLK_F_AUTO_BUF_REG")
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250621162842.337452-1-csander@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoselftests: ublk: don't take same backing file for more than one ublk devices
Ming Lei [Mon, 23 Jun 2025 01:19:27 +0000 (09:19 +0800)]
selftests: ublk: don't take same backing file for more than one ublk devices

Don't use same backing file for more than one ublk devices, and avoid
concurrent write on same file from more ublk disks.

Fixes: 8ccebc19ee3d ("selftests: ublk: support UBLK_F_AUTO_BUF_REG")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250623011934.741788-3-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agoublk: build batch from IOs in same io_ring_ctx and io task
Ming Lei [Wed, 25 Jun 2025 02:25:54 +0000 (10:25 +0800)]
ublk: build batch from IOs in same io_ring_ctx and io task

ublk_queue_cmd_list() dispatches the whole batch list by scheduling task
work via the tail request's io_uring_cmd, this way is fine even though
more than one io_ring_ctx are involved for this batch since it is just
one running context.

However, the task work handler ublk_cmd_list_tw_cb() takes `issue_flags`
of tail uring_cmd's io_ring_ctx for completing all commands. This way is
wrong if any uring_cmd is issued from different io_ring_ctx.

Fixes it by always building batch IOs from same io_ring_ctx and io task
because ublk_dispatch_req() does validate task context, and IO needs to
be aborted in case of running from fallback task work context.

For typical per-queue or per-io daemon implementation, this way shouldn't
make difference from performance viewpoint, because single io_ring_ctx is
taken in each daemon for normal use case.

Fixes: d796cea7b9f3 ("ublk: implement ->queue_rqs()")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250625022554.883571-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 weeks agobnxt: properly flush XDP redirect lists
Yan Zhai [Mon, 23 Jun 2025 16:06:38 +0000 (09:06 -0700)]
bnxt: properly flush XDP redirect lists

We encountered following crash when testing a XDP_REDIRECT feature
in production:

[56251.579676] list_add corruption. next->prev should be prev (ffff93120dd40f30), but was ffffb301ef3a6740. (next=ffff93120dd
40f30).
[56251.601413] ------------[ cut here ]------------
[56251.611357] kernel BUG at lib/list_debug.c:29!
[56251.621082] Oops: invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
[56251.632073] CPU: 111 UID: 0 PID: 0 Comm: swapper/111 Kdump: loaded Tainted: P           O       6.12.33-cloudflare-2025.6.
3 #1
[56251.653155] Tainted: [P]=PROPRIETARY_MODULE, [O]=OOT_MODULE
[56251.663877] Hardware name: MiTAC GC68B-B8032-G11P6-GPU/S8032GM-HE-CFR, BIOS V7.020.B10-sig 01/22/2025
[56251.682626] RIP: 0010:__list_add_valid_or_report+0x4b/0xa0
[56251.693203] Code: 0e 48 c7 c7 68 e7 d9 97 e8 42 16 fe ff 0f 0b 48 8b 52 08 48 39 c2 74 14 48 89 f1 48 c7 c7 90 e7 d9 97 48
 89 c6 e8 25 16 fe ff <0f> 0b 4c 8b 02 49 39 f0 74 14 48 89 d1 48 c7 c7 e8 e7 d9 97 4c 89
[56251.725811] RSP: 0018:ffff93120dd40b80 EFLAGS: 00010246
[56251.736094] RAX: 0000000000000075 RBX: ffffb301e6bba9d8 RCX: 0000000000000000
[56251.748260] RDX: 0000000000000000 RSI: ffff9149afda0b80 RDI: ffff9149afda0b80
[56251.760349] RBP: ffff9131e49c8000 R08: 0000000000000000 R09: ffff93120dd40a18
[56251.772382] R10: ffff9159cf2ce1a8 R11: 0000000000000003 R12: ffff911a80850000
[56251.784364] R13: ffff93120fbc7000 R14: 0000000000000010 R15: ffff9139e7510e40
[56251.796278] FS:  0000000000000000(0000) GS:ffff9149afd80000(0000) knlGS:0000000000000000
[56251.809133] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[56251.819561] CR2: 00007f5e85e6f300 CR3: 00000038b85e2006 CR4: 0000000000770ef0
[56251.831365] PKRU: 55555554
[56251.838653] Call Trace:
[56251.845560]  <IRQ>
[56251.851943]  cpu_map_enqueue.cold+0x5/0xa
[56251.860243]  xdp_do_redirect+0x2d9/0x480
[56251.868388]  bnxt_rx_xdp+0x1d8/0x4c0 [bnxt_en]
[56251.877028]  bnxt_rx_pkt+0x5f7/0x19b0 [bnxt_en]
[56251.885665]  ? cpu_max_write+0x1e/0x100
[56251.893510]  ? srso_alias_return_thunk+0x5/0xfbef5
[56251.902276]  __bnxt_poll_work+0x190/0x340 [bnxt_en]
[56251.911058]  bnxt_poll+0xab/0x1b0 [bnxt_en]
[56251.919041]  ? srso_alias_return_thunk+0x5/0xfbef5
[56251.927568]  ? srso_alias_return_thunk+0x5/0xfbef5
[56251.935958]  ? srso_alias_return_thunk+0x5/0xfbef5
[56251.944250]  __napi_poll+0x2b/0x160
[56251.951155]  bpf_trampoline_6442548651+0x79/0x123
[56251.959262]  __napi_poll+0x5/0x160
[56251.966037]  net_rx_action+0x3d2/0x880
[56251.973133]  ? srso_alias_return_thunk+0x5/0xfbef5
[56251.981265]  ? srso_alias_return_thunk+0x5/0xfbef5
[56251.989262]  ? __hrtimer_run_queues+0x162/0x2a0
[56251.996967]  ? srso_alias_return_thunk+0x5/0xfbef5
[56252.004875]  ? srso_alias_return_thunk+0x5/0xfbef5
[56252.012673]  ? bnxt_msix+0x62/0x70 [bnxt_en]
[56252.019903]  handle_softirqs+0xcf/0x270
[56252.026650]  irq_exit_rcu+0x67/0x90
[56252.032933]  common_interrupt+0x85/0xa0
[56252.039498]  </IRQ>
[56252.044246]  <TASK>
[56252.048935]  asm_common_interrupt+0x26/0x40
[56252.055727] RIP: 0010:cpuidle_enter_state+0xb8/0x420
[56252.063305] Code: dc 01 00 00 e8 f9 79 3b ff e8 64 f7 ff ff 49 89 c5 0f 1f 44 00 00 31 ff e8 a5 32 3a ff 45 84 ff 0f 85 ae
 01 00 00 fb 45 85 f6 <0f> 88 88 01 00 00 48 8b 04 24 49 63 ce 4c 89 ea 48 6b f1 68 48 29
[56252.088911] RSP: 0018:ffff93120c97fe98 EFLAGS: 00000202
[56252.096912] RAX: ffff9149afd80000 RBX: ffff9141d3a72800 RCX: 0000000000000000
[56252.106844] RDX: 00003329176c6b98 RSI: ffffffe36db3fdc7 RDI: 0000000000000000
[56252.116733] RBP: 0000000000000002 R08: 0000000000000002 R09: 000000000000004e
[56252.126652] R10: ffff9149afdb30c4 R11: 071c71c71c71c71c R12: ffffffff985ff860
[56252.136637] R13: 00003329176c6b98 R14: 0000000000000002 R15: 0000000000000000
[56252.146667]  ? cpuidle_enter_state+0xab/0x420
[56252.153909]  cpuidle_enter+0x2d/0x40
[56252.160360]  do_idle+0x176/0x1c0
[56252.166456]  cpu_startup_entry+0x29/0x30
[56252.173248]  start_secondary+0xf7/0x100
[56252.179941]  common_startup_64+0x13e/0x141
[56252.186886]  </TASK>

From the crash dump, we found that the cpu_map_flush_list inside
redirect info is partially corrupted: its list_head->next points to
itself, but list_head->prev points to a valid list of unflushed bq
entries.

This turned out to be a result of missed XDP flush on redirect lists. By
digging in the actual source code, we found that
commit 7f0a168b0441 ("bnxt_en: Add completion ring pointer in TX and RX
ring structures") incorrectly overwrites the event mask for XDP_REDIRECT
in bnxt_rx_xdp. We can stably reproduce this crash by returning XDP_TX
and XDP_REDIRECT randomly for incoming packets in a naive XDP program.
Properly propagate the XDP_REDIRECT events back fixes the crash.

Fixes: a7559bc8c17c ("bnxt: support transmit and free of aggregation buffers")
Tested-by: Andrew Rzeznik <arzeznik@cloudflare.com>
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Reviewed-by: Andy Gospodarek <gospo@broadcom.com>
Link: https://patch.msgid.link/aFl7jpCNzscumuN2@debian.debian
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agoMerge tag 'selinux-pr-20250624' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Wed, 25 Jun 2025 00:20:43 +0000 (17:20 -0700)]
Merge tag 'selinux-pr-20250624' of git://git./linux/kernel/git/pcmoore/selinux

Pull selinux fix from Paul Moore:
 "Another small SELinux patch to fix a problem seen by the dracut-ng
  folks during early boot when SELinux is enabled, but the policy has
  yet to be loaded"

* tag 'selinux-pr-20250624' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux:
  selinux: change security_compute_sid to return the ssid or tsid on match

3 weeks agovsock/uapi: fix linux/vm_sockets.h userspace compilation errors
Stefano Garzarella [Mon, 23 Jun 2025 10:00:53 +0000 (12:00 +0200)]
vsock/uapi: fix linux/vm_sockets.h userspace compilation errors

If a userspace application just include <linux/vm_sockets.h> will fail
to build with the following errors:

    /usr/include/linux/vm_sockets.h:182:39: error: invalid application of ‘sizeof’ to incomplete type ‘struct sockaddr’
      182 |         unsigned char svm_zero[sizeof(struct sockaddr) -
          |                                       ^~~~~~
    /usr/include/linux/vm_sockets.h:183:39: error: ‘sa_family_t’ undeclared here (not in a function)
      183 |                                sizeof(sa_family_t) -
          |

Include <sys/socket.h> for userspace (guarded by ifndef __KERNEL__)
where `struct sockaddr` and `sa_family_t` are defined.
We already do something similar in <linux/mptcp.h> and <linux/if.h>.

Fixes: d021c344051a ("VSOCK: Introduce VM Sockets")
Reported-by: Daan De Meyer <daan.j.demeyer@gmail.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://patch.msgid.link/20250623100053.40979-1-sgarzare@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
4 weeks agobcachefs: fix bch2_journal_keys_peek_prev_min() underflow
Kent Overstreet [Tue, 24 Jun 2025 20:04:32 +0000 (16:04 -0400)]
bcachefs: fix bch2_journal_keys_peek_prev_min() underflow

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
4 weeks agobcachefs: Use wait_on_allocator() when allocating journal
Kent Overstreet [Tue, 24 Jun 2025 21:53:00 +0000 (17:53 -0400)]
bcachefs: Use wait_on_allocator() when allocating journal

wait_on_allocator() emits debug info when we hang trying to allocate.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
4 weeks agobcachefs: Check for bad write buffer key when moving from journal
Kent Overstreet [Mon, 23 Jun 2025 22:42:42 +0000 (18:42 -0400)]
bcachefs: Check for bad write buffer key when moving from journal

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
4 weeks agobcachefs: Don't unlock the trans if ret doesn't match BCH_ERR_operation_blocked
Alan Huang [Tue, 24 Jun 2025 19:10:27 +0000 (03:10 +0800)]
bcachefs: Don't unlock the trans if ret doesn't match BCH_ERR_operation_blocked

Reported-by: syzbot+d540192e763531d307ff@syzkaller.appspotmail.com
Signed-off-by: Alan Huang <mmpgouride@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
4 weeks agodrm/xe/hwmon: Fix xe_hwmon_power_max_write
Karthik Poosa [Tue, 17 Jun 2025 12:00:30 +0000 (17:30 +0530)]
drm/xe/hwmon: Fix xe_hwmon_power_max_write

Prevent other bits of mailbox power limit from being overwritten with 0.
This issue was due to a missing read and modify of current power limit,
before setting a requested mailbox power limit, which is added in this
patch.

v2:
 - Improve commit message. (Anshuman)

v3:
 - Rebase.
 - Rephrase commit message. (Riana)
 - Add read-modify-write variant of xe_hwmon_pcode_write_power_limit()
   i.e. xe_hwmon_pcode_rmw_power_limit(). (Badal)
 - Use xe_hwmon_pcode_rmw_power_limit() to set mailbox power limits.
 - Remove xe_hwmon_pcode_write_power_limit() as all mailbox power limits
   writes use xe_hwmon_pcode_rmw_power_limit() only.

v4:
 - Use PWR_LIM in place of (PWR_LIM_EN | PWR_LIM_VAL) wherever
   applicable. (Riana)

Fixes: 25a2aa779fc3 ("drm/xe/hwmon: Add support to manage power limits though mailbox")
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Signed-off-by: Karthik Poosa <karthik.poosa@intel.com>
Reviewed-by: Badal Nilawar <badal.nilawar@intel.com>
Link: https://lore.kernel.org/r/20250617120030.612819-1-karthik.poosa@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
(cherry picked from commit 8aa7306631f088881759398972d503757cf0c901)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
4 weeks agodrm/xe/display: Add check for alloc_ordered_workqueue()
Haoxiang Li [Fri, 16 May 2025 12:16:55 +0000 (15:16 +0300)]
drm/xe/display: Add check for alloc_ordered_workqueue()

Add check for the return value of alloc_ordered_workqueue()
in xe_display_create() to catch potential exception.

Fixes: 44e694958b95 ("drm/xe/display: Implement display support")
Cc: stable@vger.kernel.org
Signed-off-by: Haoxiang Li <haoxiang_li2024@163.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/4ee1b0e5d1626ce1dde2e82af05c2edaed50c3aa.1747397638.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
(cherry picked from commit 5b62d63395d5b7d4094e7cd380bccae4b25415cb)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
4 weeks agospi: spi-cadence-quadspi: Fix pm runtime unbalance
Khairul Anuar Romli [Mon, 16 Jun 2025 01:13:53 +0000 (09:13 +0800)]
spi: spi-cadence-quadspi: Fix pm runtime unbalance

Having PM put sync in remove function is causing PM underflow during
remove operation. This is caused by the function, runtime_pm_get_sync,
not being called anywhere during the op. Ensure that calls to
pm_runtime_enable()/pm_runtime_disable() and
pm_runtime_get_sync()/pm_runtime_put_sync() match.

echo 108d2000.spi > /sys/bus/platform/drivers/cadence-qspi/unbind
[   49.644256] Deleting MTD partitions on "108d2000.spi.0":
[   49.649575] Deleting u-boot MTD partition
[   49.684087] Deleting root MTD partition
[   49.724188] cadence-qspi 108d2000.spi: Runtime PM usage count underflow!

Continuous bind/unbind will result in an "Unbalanced pm_runtime_enable" error.
Subsequent unbind attempts will return a "No such device" error, while bind
attempts will return a "Resource temporarily unavailable" error.

[   47.592434] cadence-qspi 108d2000.spi: Runtime PM usage count underflow!
[   49.592233] cadence-qspi 108d2000.spi: detected FIFO depth (1024) different from config (128)
[   53.232309] cadence-qspi 108d2000.spi: Runtime PM usage count underflow!
[   55.828550] cadence-qspi 108d2000.spi: detected FIFO depth (1024) different from config (128)
[   57.940627] cadence-qspi 108d2000.spi: Runtime PM usage count underflow!
[   59.912490] cadence-qspi 108d2000.spi: detected FIFO depth (1024) different from config (128)
[   61.876243] cadence-qspi 108d2000.spi: Runtime PM usage count underflow!
[   61.883000] platform 108d2000.spi: Unbalanced pm_runtime_enable!
[  532.012270] cadence-qspi 108d2000.spi: probe with driver cadence-qspi failed1

Also, change clk_disable_unprepare() to clk_disable() since continuous
bind and unbind operations will trigger a warning indicating that the clock is
already unprepared.

Fixes: 4892b374c9b7 ("mtd: spi-nor: cadence-quadspi: Add runtime PM support")
cc: stable@vger.kernel.org # 6.6+
Signed-off-by: Khairul Anuar Romli <khairul.anuar.romli@altera.com>
Reviewed-by: Matthew Gerlach <matthew.gerlach@altera.com>
Link: https://patch.msgid.link/4e7a4b8aba300e629b45a04f90bddf665fbdb335.1749601877.git.khairul.anuar.romli@altera.com
Signed-off-by: Mark Brown <broonie@kernel.org>
4 weeks agoASoC: SOF: Intel: hda: Use devm_kstrdup() to avoid memleak.
Tamura Dai [Sun, 15 Jun 2025 23:55:48 +0000 (08:55 +0900)]
ASoC: SOF: Intel: hda: Use devm_kstrdup() to avoid memleak.

sof_pdata->tplg_filename can have address allocated by kstrdup()
and can be overwritten. Memory leak was detected with kmemleak:

unreferenced object 0xffff88812391ff60 (size 16):
  comm "kworker/4:1", pid 161, jiffies 4294802931
  hex dump (first 16 bytes):
    73 6f 66 2d 68 64 61 2d 67 65 6e 65 72 69 63 00  sof-hda-generic.
  backtrace (crc 4bf1675c):
    __kmalloc_node_track_caller_noprof+0x49c/0x6b0
    kstrdup+0x46/0xc0
    hda_machine_select.cold+0x1de/0x12cf [snd_sof_intel_hda_generic]
    sof_init_environment+0x16f/0xb50 [snd_sof]
    sof_probe_continue+0x45/0x7c0 [snd_sof]
    sof_probe_work+0x1e/0x40 [snd_sof]
    process_one_work+0x894/0x14b0
    worker_thread+0x5e5/0xfb0
    kthread+0x39d/0x760
    ret_from_fork+0x31/0x70
    ret_from_fork_asm+0x1a/0x30

Signed-off-by: Tamura Dai <kirinode0@gmail.com>
Link: https://patch.msgid.link/20250615235548.8591-1-kirinode0@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
4 weeks agodrm/amd/display: Add sanity checks for drm_edid_raw()
Takashi Iwai [Mon, 16 Jun 2025 16:08:41 +0000 (18:08 +0200)]
drm/amd/display: Add sanity checks for drm_edid_raw()

When EDID is retrieved via drm_edid_raw(), it doesn't guarantee to
return proper EDID bytes the caller wants: it may be either NULL (that
leads to an Oops) or with too long bytes over the fixed size raw_edid
array (that may lead to memory corruption).  The latter was reported
actually when connected with a bad adapter.

Add sanity checks for drm_edid_raw() to address the above corner
cases, and return EDID_BAD_INPUT accordingly.

Fixes: 48edb2a4256e ("drm/amd/display: switch amdgpu_dm_connector to use struct drm_edid")
Link: https://bugzilla.suse.com/show_bug.cgi?id=1236415
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 648d3f4d209725d51900d6a3ed46b7b600140cdf)
Cc: stable@vger.kernel.org
4 weeks agodrm/amd/display: Fix AMDGPU_MAX_BL_LEVEL value
Mario Limonciello [Mon, 23 Jun 2025 17:11:13 +0000 (12:11 -0500)]
drm/amd/display: Fix AMDGPU_MAX_BL_LEVEL value

[Why]
commit 16dc8bc27c2a ("drm/amd/display: Export full brightness range to
userspace") adjusted the brightness range to scale to larger values, but
missed updating AMDGPU_MAX_BL_LEVEL which is needed to make sure that
scaling works properly with custom brightness curves.

[How]
As the change for max brightness of 0xFFFF only applies to devices
supporting DC, use existing DC define MAX_BACKLIGHT_LEVEL.

Fixes: 16dc8bc27c2a ("drm/amd/display: Export full brightness range to userspace")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250623171114.1156451-1-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 5b852044eb0d3e1f1c946d32e05fcb068e0a20a0)
Cc: stable@vger.kernel.org
4 weeks agodrm/amdgpu/sdma7: add ucode version checks for userq support
Alex Deucher [Fri, 20 Jun 2025 15:39:22 +0000 (11:39 -0400)]
drm/amdgpu/sdma7: add ucode version checks for userq support

SDMA 7.0.0/1: 7836028

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 8c011408ed842dfccdd50a90a9cf6bccdb85cc0e)

4 weeks agodrm/amdgpu/sdma6: add ucode version checks for userq support
Alex Deucher [Thu, 19 Jun 2025 21:56:29 +0000 (17:56 -0400)]
drm/amdgpu/sdma6: add ucode version checks for userq support

SDMA 6.0.0 version 24
SDMA 6.0.2 version 21
SDMA 6.0.3 version 25

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit e8cca30d8b34f1c4101c237914c53068d4a55e73)

4 weeks agodrm/amd: Adjust output for discovery error handling
Mario Limonciello [Tue, 17 Jun 2025 18:30:52 +0000 (13:30 -0500)]
drm/amd: Adjust output for discovery error handling

commit 017fbb6690c2 ("drm/amdgpu/discovery: check ip_discovery fw file
available") added support for reading an amdgpu IP discovery bin file
for some specific products. If it's not found then it will fallback to
hardcoded values. However if it's not found there is also a lot of noise
about missing files and errors.

Adjust the error handling to decrease most messages to DEBUG and to show
users less about missing files.

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reported-by: Marcus Seyfarth <m.seyfarth@gmail.com>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4312
Tested-by: Marcus Seyfarth <m.seyfarth@gmail.com>
Fixes: 017fbb6690c2 ("drm/amdgpu/discovery: check ip_discovery fw file available")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250617183052.1692059-1-superm1@kernel.org
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 49f1f9f6c3c9febf8ba93f94a8d9c8d03e1ea0a1)

4 weeks agodrm/amdgpu/mes: add compatibility checks for set_hw_resource_1
Alex Deucher [Tue, 20 May 2025 14:02:14 +0000 (10:02 -0400)]
drm/amdgpu/mes: add compatibility checks for set_hw_resource_1

Seems some older MES firmware versions do not properly support
this packet.  Add back some the compatibility checks.

v2: switch to fw version check (Shaoyun)

Fixes: f81cd793119e ("drm/amd/amdgpu: Fix MES init sequence")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4295
Cc: Shaoyun Liu <shaoyun.liu@amd.com>
Reviewed-by: shaoyun.liu <shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 0180e0a5dd5c6ff118043ee42dbbbddaf881f283)
Cc: stable@vger.kernel.org
4 weeks agodrm/amdgpu/gfx9: Add Cleaner Shader Support for GFX9.x GPUs
Srinivasan Shanmugam [Thu, 12 Jun 2025 14:41:14 +0000 (20:11 +0530)]
drm/amdgpu/gfx9: Add Cleaner Shader Support for GFX9.x GPUs

Enable the cleaner shader for other GFX9.x series of GPUs to provide
data isolation between GPU workloads. The cleaner shader is responsible
for clearing the Local Data Store (LDS), Vector General Purpose
Registers (VGPRs), and Scalar General Purpose Registers (SGPRs), which
helps prevent data leakage and ensures accurate computation results.

This update extends cleaner shader support to GFX9.x GPUs, previously
available for GFX9.4.2. It enhances security by clearing GPU memory
between processes and maintains a consistent GPU state across KGD and
KFD workloads.

Cc: Manu Rastogi <manu.rastogi@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 99808926d0ea6234a89e35240a7cb088368de9e1)

4 weeks agouserns and mnt_idmap leak in open_tree_attr(2)
Al Viro [Tue, 24 Jun 2025 14:25:04 +0000 (10:25 -0400)]
userns and mnt_idmap leak in open_tree_attr(2)

Once want_mount_setattr() has returned a positive, it does require
finish_mount_kattr() to release ->mnt_userns.  Failing do_mount_setattr()
does not change that.

As the result, we can end up leaking userns and possibly mnt_idmap as
well.

Fixes: c4a16820d901 ("fs: add open_tree_attr()")
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
4 weeks agowifi: mac80211: finish link init before RCU publish
Johannes Berg [Tue, 24 Jun 2025 11:07:49 +0000 (13:07 +0200)]
wifi: mac80211: finish link init before RCU publish

Since the link/conf pointers can be accessed without any
protection other than RCU, make sure the data is actually
set up before publishing the structures.

Fixes: b2e8434f1829 ("wifi: mac80211: set up/tear down client vif links properly")
Link: https://patch.msgid.link/20250624130749.9a308b713c74.I4a80f5eead112a38730939ea591d2e275c721256@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>