ALSA: hda/realtek - Fix inactive headset mic jack for ASUS Vivobook 15 X1504VAP

When the headset is connected, there is no automatic switching of the
capture source - you can only manually select the headset microphone
in pavucontrol.

This patch fixes/activates the inactive microphone of the headset.

Signed-off-by: Vasiliy Kovalev <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek: Support mute LED on HP Laptop 14-dq2xxx

The mute LED on this HP laptop uses ALC236 and requires a quirk to function. This patch enables the existing quirk for the device.

Signed-off-by: Maximilien Perreault <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek: Enable Mute Led for HP Victus 15-fb1xxx

The mute led is controlled by ALC245. This patch enables the already
existing quirk for this device.

Signed-off-by: Adam Queler <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek: extend quirks for Clevo V5[46]0

The mic in those laptops suffers too high gain resulting in mostly (fan
or else) noise being recorded. In addition to the existing fixup about
mic detection, apply also limiting its boost. While at it, extend the
quirk to also V5[46]0TNE models, which have the same issue.

Signed-off-by: Marek Marczykowski-Górecki <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda: add HDMI codec ID for Intel PTL

Add HDMI codec ID for Intel Panther Lake platform.

Signed-off-by: Kai Vehmanen <[email protected]>
Reviewed-by: Péter Ujfalusi <[email protected]>
Reviewed-by: Ranjani Sridharan <[email protected]>
Signed-off-by: Bard Liao <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek: add patch for internal mic in Lenovo V145

Lenovo V145 is having phase inverted dmic but simply applying inverted
dmic fixups does not work. Chaining up verb fixes for ALC283 enables
inverting dmic fixup to work properly.

Signed-off-by: Terry Cheong <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/conexant: Add pincfg quirk to enable top speakers on Sirius devices

The Sirius notebooks have two sets of speakers 0x17 (sides) and
0x1d (top center). The side speakers are active by default but
the top speakers aren't.

This patch provides a pincfg quirk to activate the top speakers.

Signed-off-by: Christoffer Sandberg <[email protected]>
Signed-off-by: Werner Sembach <[email protected]>
Cc: [email protected]
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda: hda_component: Fix mutex crash if nothing ever binds

Move the initialization of parent->mutex into
hda_component_manager_init() so that it is always valid.

In hda_component_manager_bind() do not clear the parent information.
Only zero-fill the per-component data ready for it to be filled in
by the components as they bind.

Previously parent->mutex was being initialized only in
hda_component_manager_bind(). This meant that it was only
initialized if all components appeared and there was a bind callback.
If there wasn't a bind the mutex object was not valid when the
Realtek driver called any of the other functions.

Signed-off-by: Richard Fitzgerald <[email protected]>
Fixes: 047b9cbbaa8e ("ALSA: hda: hda_component: Protect shared data with a mutex")
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek: support HP Pavilion Aero 13-bg0xxx Mute LED

This patch adds the HP Pavilion Aero 13 (13-bg0xxx) (year 2024) to list of
quirks for keyboard LED mute indication.

The laptop has two LEDs (one for speaker and one for mic mute). The
pre-existing quirk ALC245_FIXUP_HP_X360_MUTE_LEDS chains both the quirk for
mic and speaker mute.

Tested on 6.11.0-rc4 with the aforementioned laptop.

Signed-off-by: Hendrik Borghorst <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek: Fix the speaker output on Samsung Galaxy Book3 Ultra

144d:c1cc requires the same workaround to enable the speaker amp
as other Samsung models with the ALC298 codec.

Signed-off-by: YOUNGJIN JOO <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

Merge tag 'asoc-fix-v6.11-rc3' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v6.11

A relatively large collection of fixes here, all driver specific and
none of them particularly major, plus one MAINTAINERS update.  There's
been a bunch of work on module autoloading from several people.

ASoC: cs-amp-lib: Ignore empty UEFI calibration entries

If the timestamp of a calibration entry is 0 it is an unused entry and
must be ignored.

Some end-products reserve EFI space for calibration entries by shipping
with a zero-filled EFI file. When searching the file for calibration
data the driver must skip the empty entries. The timestamp of a valid
entry is always non-zero.

Signed-off-by: Richard Fitzgerald <[email protected]>
Fixes: 1cad8725f2b9 ("ASoC: cs-amp-lib: Add helpers for factory calibration data")
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>

ASoC: cs-amp-lib-test: Force test calibration blob entries to be valid

For a normal calibration blob the calTarget values must be non-zero and
unique, and the calTime values must be non-zero. Don't rely on
get_random_bytes() to be random enough to guarantee this. Force the
calTarget and calTime values to be valid while retaining randomness
in the values.

Signed-off-by: Richard Fitzgerald <[email protected]>
Fixes: 177862317a98 ("ASoC: cs-amp-lib: Add KUnit test for calibration helpers")
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>

ALSA: hda/realtek - FIxed ALC285 headphone no sound

Dell platform with ALC215 ALC285 ALC289 ALC225 ALC295 ALC299, plug
headphone or headset.
It had a chance to get no sound from headphone.
Replace depop procedure will solve this issue.

Signed-off-by: Kailang Yang <[email protected]>
Link: https://lore.kernel.org/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek - Fixed ALC256 headphone no sound

Dell platform, plug headphone or headset, it had a chance to get no
sound from headphone.
Replace depop procedure will solve this issue.

Signed-off-by: Kailang Yang <[email protected]>
Link: https://lore.kernel.org/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ASoC: allow module autoloading for table board_ids

Add MODULE_DEVICE_TABLE(), so modules could be properly
autoloaded based on the alias from platform_device_id table.

Signed-off-by: Hongbo Li <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>

ASoC: allow module autoloading for table db1200_pids

Add MODULE_DEVICE_TABLE(), so modules could be properly
autoloaded based on the alias from platform_device_id table.

Signed-off-by: Hongbo Li <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>

ALSA: hda: cs35l56: Don't use the device index as a calibration index

The HDA driver cannot assume that the order that the devices are
specified in the cirrus,dev-index matches the order of calibration
entries.

Only a calibration entry with a matching silicon id will be used.

Fixes: cfa43aaa7948 ("ALSA: hda: cs35l56: Apply amp calibration from EFI data")
Signed-off-by: Simon Trimmer <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: seq: Skip event type filtering for UMP events

UMP events don't use the event type field, hence it's invalid to apply
the filter, which may drop the events unexpectedly.
Skip the event filtering for UMP events, instead.

Fixes: 46397622a3fa ("ALSA: seq: Add UMP support")
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hda/realtek: Enable mute/micmute LEDs on HP Laptop 14-ey0xxx

HP Pavilion Plus 14-ey0xxx needs existing quirk
ALC245_FIXUP_HP_X360_MUTE_LEDS to enable its mute/micmute LEDs.

Signed-off-by: John Sweeney <[email protected]>
Cc: <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Takashi Iwai <[email protected]>

Linux 6.11-rc4

Merge tag 'driver-core-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core

Pull driver core fixes from Greg KH:
 "Here are two driver fixes for regressions from 6.11-rc1 due to the
  driver core change making a structure in a driver core callback const.

  These were missed by all testing EXCEPT for what Bart happened to be
  running, so I appreciate the fixes provided here for some
  odd/not-often-used driver subsystems that nothing else happened to
  catch.

  Both of these fixes have been in linux-next all week with no reported
  issues"

* tag 'driver-core-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  mips: sgi-ip22: Fix the build
  ARM: riscpc: ecard: Fix the build

Merge tag 'char-misc-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc

Pull char / misc fixes from Greg KH:
 "Here are some small char/misc fixes for 6.11-rc4 to resolve reported
  problems. Included in here are:

   - fastrpc revert of a change that broke userspace

   - xillybus fixes for reported issues

  Half of these have been in linux-next this week with no reported
  problems, I don't know if the last bit of xillybus driver changes made
  it in, but they are 'obviously correct' so will be safe :)"

* tag 'char-misc-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  char: xillybus: Check USB endpoints when probing device
  char: xillybus: Refine workqueue handling
  Revert "misc: fastrpc: Restrict untrusted app to attach to privileged PD"
  char: xillybus: Don't destroy workqueue from work item running on it

Merge tag 'tty-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty

Pull tty / serial fixes from Greg KH:
 "Here are some small tty and serial driver fixes for 6.11-rc4 to
  resolve some reported problems. Included in here are:

   - conmakehash.c userspace build issues

   - fsl_lpuart driver fix

   - 8250_omap revert for reported regression

   - atmel_serial rts flag fix

  All of these have been in linux-next this week with no reported
  issues"

* tag 'tty-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  Revert "serial: 8250_omap: Set the console genpd always on if no console suspend"
  tty: atmel_serial: use the correct RTS flag.
  tty: vt: conmakehash: remove non-portable code printing comment header
  tty: serial: fsl_lpuart: mark last busy before uart_add_one_port

Merge tag 'usb-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb

Pull USB / Thunderbolt driver fixes from Greg KH:
 "Here are some small USB and Thunderbolt driver fixes for 6.11-rc4 to
  resolve some reported issues. Included in here are:

   - thunderbolt driver fixes for reported problems

   - typec driver fixes

   - xhci fixes

   - new device id for ljca usb driver

  All of these have been in linux-next this week with no reported
  issues"

* tag 'usb-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  xhci: Fix Panther point NULL pointer deref at full-speed re-enumeration
  usb: misc: ljca: Add Lunar Lake ljca GPIO HID to ljca_gpio_hids[]
  Revert "usb: typec: tcpm: clear pd_event queue in PORT_RESET"
  usb: typec: ucsi: Fix the return value of ucsi_run_command()
  usb: xhci: fix duplicate stall handling in handle_tx_event()
  usb: xhci: Check for xhci->interrupters being allocated in xhci_mem_clearup()
  thunderbolt: Mark XDomain as unplugged when router is removed
  thunderbolt: Fix memory leaks in {port|retimer}_sb_regs_write()

Merge tag 'for-6.11-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull more btrfs fixes from David Sterba:
 "A more fixes. We got reports that shrinker added in 6.10 still causes
  latency spikes and the fixes don't handle all corner cases. Due to
  summer holidays we're taking a shortcut to disable it for release
  builds and will fix it in the near future.

   - only enable extent map shrinker for DEBUG builds, temporary quick
     fix to avoid latency spikes for regular builds

   - update target inode's ctime on unlink, mandated by POSIX

   - properly take lock to read/update block group's zoned variables

   - add counted_by() annotations"

* tag 'for-6.11-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: only enable extent map shrinker for DEBUG builds
  btrfs: zoned: properly take lock to read/update block group's zoned variables
  btrfs: tree-checker: add dev extent item checks
  btrfs: update target inode's ctime on unlink
  btrfs: send: annotate struct name_cache_entry with __counted_by()

fuse: Initialize beyond-EOF page contents before setting uptodate

fuse_notify_store(), unlike fuse_do_readpage(), does not enable page
zeroing (because it can be used to change partial page contents).

So fuse_notify_store() must be more careful to fully initialize page
contents (including parts of the page that are beyond end-of-file)
before marking the page uptodate.

The current code can leave beyond-EOF page contents uninitialized, which
makes these uninitialized page contents visible to userspace via mmap().

This is an information leak, but only affects systems which do not
enable init-on-alloc (via CONFIG_INIT_ON_ALLOC_DEFAULT_ON=y or the
corresponding kernel command line parameter).

Link: https://bugs.chromium.org/p/project-zero/issues/detail?id=2574
Cc: [email protected]
Fixes: a1d75f258230 ("fuse: add store request")
Signed-off-by: Jann Horn <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

Merge tag 'mm-hotfixes-stable-2024-08-17-19-34' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
 "16 hotfixes. All except one are for MM. 10 of these are cc:stable and
  the others pertain to post-6.10 issues.

  As usual with these merges, singletons and doubletons all over the
  place, no identifiable-by-me theme. Please see the lovingly curated
  changelogs to get the skinny"

* tag 'mm-hotfixes-stable-2024-08-17-19-34' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  mm/migrate: fix deadlock in migrate_pages_batch() on large folios
  alloc_tag: mark pages reserved during CMA activation as not tagged
  alloc_tag: introduce clear_page_tag_ref() helper function
  crash: fix riscv64 crash memory reserve dead loop
  selftests: memfd_secret: don't build memfd_secret test on unsupported arches
  mm: fix endless reclaim on machines with unaccepted memory
  selftests/mm: compaction_test: fix off by one in check_compaction()
  mm/numa: no task_numa_fault() call if PMD is changed
  mm/numa: no task_numa_fault() call if PTE is changed
  mm/vmalloc: fix page mapping if vm_area_alloc_pages() with high order fallback to order 0
  mm/memory-failure: use raw_spinlock_t in struct memory_failure_cpu
  mm: don't account memmap per-node
  mm: add system wide stats items category
  mm: don't account memmap on failure
  mm/hugetlb: fix hugetlb vs. core-mm PT locking
  mseal: fix is_madv_discard()

Merge tag 'powerpc-6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:

 - Fix crashes on 85xx with some configs since the recent hugepd rework.

 - Fix boot warning with hugepages and CONFIG_DEBUG_VIRTUAL on some
   platforms.

 - Don't enable offline cores when changing SMT modes, to match existing
   userspace behaviour.

Thanks to Christophe Leroy, Dr. David Alan Gilbert, Guenter Roeck, Nysal
Jan K.A, Shrikanth Hegde, Thomas Gleixner, and Tyrel Datwyler.

* tag 'powerpc-6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/topology: Check if a core is online
  cpu/SMT: Enable SMT only if a core is online
  powerpc/mm: Fix boot warning with hugepages and CONFIG_DEBUG_VIRTUAL
  powerpc/mm: Fix size of allocated PGDIR
  soc: fsl: qbman: remove unused struct 'cgr_comp'

Merge tag 'v6.11-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6

Pull smb client fixes from Steve French:

 - fix for clang warning - additional null check

 - fix for cached write with posix locks

 - flexible structure fix

* tag 'v6.11-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
  smb: smb2pdu.h: Use static_assert() to check struct sizes
  smb3: fix lock breakage for cached writes
  smb/client: avoid possible NULL dereference in cifs_free_subrequest()

Merge tag 'i2c-for-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux

Pull i2c fixes from Wolfram Sang:
 "I2C core fix replacing IS_ENABLED() with IS_REACHABLE()

  For host drivers, there are two fixes:

   - Tegra I2C Controller: Addresses a potential double-locking issue
     during probe. ACPI devices are not IRQ-safe when invoking runtime
     suspend and resume functions, so the irq_safe flag should not be
     set.

   - Qualcomm GENI I2C Controller: Fixes an oversight in the exit path
     of the runtime_resume() function, which was missed in the previous
     release"

* tag 'i2c-for-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: tegra: Do not mark ACPI devices as irq safe
  i2c: Use IS_REACHABLE() for substituting empty ACPI functions
  i2c: qcom-geni: Add missing geni_icc_disable in geni_i2c_runtime_resume

Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "Two small fixes to the mpi3mr driver. One to avoid oversize
  allocations in tracing and the other to fix an uninitialized spinlock
  in the user to driver feature request code (used to trigger dumps and
  the like)"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: mpi3mr: Avoid MAX_PAGE_ORDER WARNING for buffer allocations
  scsi: mpi3mr: Add missing spin_lock_init() for mrioc->trigger_lock

Merge tag 'xfs-6.11-fixes-3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull xfs fixes from Chandan Babu:

 - Check for presence of only 'attr' feature before scrubbing an inode's
   attribute fork.

 - Restore the behaviour of setting AIL thread to TASK_INTERRUPTIBLE for
   long (i.e. 50ms) sleep durations to prevent high load averages.

 - Do not allow users to change the realtime flag of a file unless the
   datadev and rtdev both support fsdax access modes.

* tag 'xfs-6.11-fixes-3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: conditionally allow FS_XFLAG_REALTIME changes if S_DAX is set
  xfs: revert AIL TASK_KILLABLE threshold
  xfs: attr forks require attr, not attr2

Merge tag 'bcachefs-2024-08-16' of git://evilpiepirate.org/bcachefs

Pull bcachefs fixes from Kent OverstreetL

 - New on disk format version, bcachefs_metadata_version_disk_accounting_inum

   This adds one more disk accounting counter, which counts disk usage
   and number of extents per inode number. This lets us track
   fragmentation, for implementing defragmentation later, and it also
   counts disk usage per inode in all snapshots, which will be a useful
   thing to expose to users.

 - One performance issue we've observed is threads spinning when they
   should be waiting for dirty keys in the key cache to be flushed by
   journal reclaim, so we now have hysteresis for the waiting thread, as
   well as improving the tracepoint and a new time_stat, for tracking
   time blocked waiting on key cache flushing.

... and various assorted smaller fixes.

* tag 'bcachefs-2024-08-16' of git://evilpiepirate.org/bcachefs:
  bcachefs: Fix locking in __bch2_trans_mark_dev_sb()
  bcachefs: fix incorrect i_state usage
  bcachefs: avoid overflowing LRU_TIME_BITS for cached data lru
  bcachefs: Fix forgetting to pass trans to fsck_err()
  bcachefs: Increase size of cuckoo hash table on too many rehashes
  bcachefs: bcachefs_metadata_version_disk_accounting_inum
  bcachefs: Kill __bch2_accounting_mem_mod()
  bcachefs: Make bkey_fsck_err() a wrapper around fsck_err()
  bcachefs: Fix warning in __bch2_fsck_err() for trans not passed in
  bcachefs: Add a time_stat for blocked on key cache flush
  bcachefs: Improve trans_blocked_journal_reclaim tracepoint
  bcachefs: Add hysteresis to waiting on btree key cache flush
  lib/generic-radix-tree.c: Fix rare race in __genradix_ptr_alloc()
  bcachefs: Convert for_each_btree_node() to lockrestart_do()
  bcachefs: Add missing downgrade table entry
  bcachefs: disk accounting: ignore unknown types
  bcachefs: bch2_accounting_invalid() fixup
  bcachefs: Fix bch2_trigger_alloc when upgrading from old versions
  bcachefs: delete faulty fastpath in bch2_btree_path_traverse_cached()

bcachefs: Fix locking in __bch2_trans_mark_dev_sb()

We run this in full RW mode now, so we have to guard against the
superblock buffer being reallocated.

Signed-off-by: Kent Overstreet <[email protected]>

Merge tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull memcg-v1 fix from Al Viro:
 "memcg_write_event_control() oops fix"

* tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  memcg_write_event_control(): fix a user-triggerable oops

Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

Pull arm64 fixes from Catalin Marinas:

 - Fix the arm64 __get_mem_asm() to use the _ASM_EXTABLE_##type##ACCESS()
   macro instead of the *_ERR() one in order to avoid writing -EFAULT to
   the value register in case of a fault

 - Initialise all elements of the acpi_early_node_map[] to NUMA_NO_NODE.
   Prior to this fix, only the first element was initialised

 - Move the KASAN random tag seed initialisation after the per-CPU areas
   have been initialised (prng_state is __percpu)

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: Fix KASAN random tag seed initialization
  arm64: ACPI: NUMA: initialize all values of acpi_early_node_map to NUMA_NO_NODE
  arm64: uaccess: correct thinko in __get_mem_asm()

Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux

Pull clk fix from Stephen Boyd:
 "One fix for the new T-Head TH1520 clk driver that marks a bus clk
  critical so that it isn't turned off during late init which breaks
  emmc-sdio"

* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
  clk: thead: fix dependency on clk_ignore_unused

Merge tag 'block-6.11-20240824' of git://git.kernel.dk/linux

Pull block fixes from Jens Axboe:

 - Fix corruption issues with s390/dasd (Eric, Stefan)

 - Fix a misuse of non irq locking grab of a lock (Li)

 - MD pull request with a single data corruption fix for raid1 (Yu)

* tag 'block-6.11-20240824' of git://git.kernel.dk/linux:
  block: Fix lockdep warning in blk_mq_mark_tag_wait
  md/raid1: Fix data corruption for degraded array with slow disk
  s390/dasd: fix error recovery leading to data corruption on ESE devices
  s390/dasd: Remove DMA alignment

Merge tag 'io_uring-6.11-20240824' of git://git.kernel.dk/linux

Pull io_uring fixes from Jens Axboe:

 - Fix a comment in the uapi header using the wrong member name (Caleb)

 - Fix KCSAN warning for a debug check in sqpoll (me)

 - Two more NAPI tweaks (Olivier)

* tag 'io_uring-6.11-20240824' of git://git.kernel.dk/linux:
  io_uring: fix user_data field name in comment
  io_uring/sqpoll: annotate debug task == current with data_race()
  io_uring/napi: remove duplicate io_napi_entry timeout assignation
  io_uring/napi: check napi_enabled in io_napi_add() before proceeding

Merge tag 'devicetree-fixes-for-6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux

Pull devicetree fixes from Rob Herring:

 - Fix a possible (but unlikely) out-of-bounds read in interrupts
   parsing code

 - Add AT25 EEPROM "fujitsu,mb85rs256" compatible

 - Update Konrad Dybcio's email

* tag 'devicetree-fixes-for-6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
  of/irq: Prevent device address out-of-bounds read in interrupt map walk
  dt-bindings: eeprom: at25: add fujitsu,mb85rs256 compatible
  dt-bindings: Batch-update Konrad Dybcio's email

btrfs: only enable extent map shrinker for DEBUG builds

Although there are several patches improving the extent map shrinker,
there are still reports of too frequent shrinker behavior, taking too
much CPU for the kswapd process.

So let's only enable extent shrinker for now, until we got more
comprehensive understanding and a better solution.

Link: https://lore.kernel.org/linux-btrfs/[email protected]/
Link: https://lore.kernel.org/linux-btrfs/[email protected]/
Fixes: 956a17d9d050 ("btrfs: add a shrinker for extent maps")
CC: [email protected] # 6.10+
Signed-off-by: Qu Wenruo <[email protected]>
Reviewed-by: David Sterba <[email protected]>
Signed-off-by: David Sterba <[email protected]>

Merge tag 'thermal-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull thermal control fix from Rafael Wysocki:
 "Fix a Bang-bang thermal governor issue causing it to fail to reset the
  state of cooling devices if they are 'on' to start with, but the
  thermal zone temperature is always below the corresponding trip point
  (Rafael Wysocki)"

* tag 'thermal-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  thermal: gov_bang_bang: Use governor_data to reduce overhead
  thermal: gov_bang_bang: Add .manage() callback
  thermal: gov_bang_bang: Split bang_bang_control()
  thermal: gov_bang_bang: Call __thermal_cdev_update() directly

Merge tag 'acpi-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull ACPI fix from Rafael Wysocki:
 "Fix an issue related to the ACPI EC device handling that causes the
  _REG control method to be evaluated for EC operation regions that are
  not expected to be used.

  This confuses the platform firmware and provokes various types of
  misbehavior on some systems (Rafael Wysocki)"

* tag 'acpi-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: EC: Evaluate _REG outside the EC scope more carefully
  ACPICA: Add a depth argument to acpi_execute_reg_methods()
  Revert "ACPI: EC: Evaluate orphan _REG under EC device"

Merge tag 'libnvdimm-fixes-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm fix from Ira Weiny:
 "Commit f467fee48da4 ("block: move the dax flag to queue_limits") broke
  the DAX tests by skipping over the legacy pmem mapping pages case.

  Set the DAX flag in this case as well"

* tag 'libnvdimm-fixes-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  nvdimm/pmem: Set dax flag for all 'PFN_MAP' cases

io_uring: fix user_data field name in comment

io_uring_cqe's user_data field refers to `sqe->data`, but io_uring_sqe
does not have a data field. Fix the comment to say `sqe->user_data`.

Signed-off-by: Caleb Sander Mateos <[email protected]>
Link: https://github.com/axboe/liburing/pull/1206
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

Merge tag 'rust-fixes-6.11' of https://github.com/Rust-for-Linux/linux

Pull rust fixes from Miguel Ojeda:

 - Fix '-Os' Rust 1.80.0+ builds adding more intrinsics (also tweaked in
   upstream Rust for the upcoming 1.82.0).

 - Fix support for the latest version of rust-analyzer due to a change
   on rust-analyzer config file semantics (considered a fix since most
   developers use the latest version of the tool, which is the only one
   actually supported by upstream). I am discussing stability of the
   config file with upstream -- they may be able to start versioning it.

 - Fix GCC 14 builds due to '-fmin-function-alignment' not skipped for
   libclang (bindgen).

 - A couple Kconfig fixes around '{RUSTC,BINDGEN}_VERSION_TEXT' to
   suppress error messages in a foreign architecture chroot and to use a
   proper default format.

 - Clean 'rust-analyzer' target warning due to missing recursive make
   invocation mark.

 - Clean Clippy warning due to missing indentation in docs.

 - Clean LLVM 19 build warning due to removed 3dnow feature upstream.

* tag 'rust-fixes-6.11' of https://github.com/Rust-for-Linux/linux:
  rust: x86: remove `-3dnow{,a}` from target features
  kbuild: rust-analyzer: mark `rust_is_available.sh` invocation as recursive
  rust: add intrinsics to fix `-Os` builds
  kbuild: rust: skip -fmin-function-alignment in bindgen flags
  rust: Support latest version of `rust-analyzer`
  rust: macros: indent list item in `module!`'s docs
  rust: fix the default format for CONFIG_{RUSTC,BINDGEN}_VERSION_TEXT
  rust: suppress error messages from CONFIG_{RUSTC,BINDGEN}_VERSION_TEXT

Merge tag 'riscv-for-linus-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux

Pull RISC-V fixes from Palmer Dabbelt:

 - reintroduce the text patching global icache flush

 - fix syscall entry code to correctly initialize a0, which manifested
   as a strace bug

 - XIP kernels now map the entire kernel, which fixes boot under at
   least DEBUG_VIRTUAL=y

 - initialize all nodes in the acpi_early_node_map initializer

 - fix OOB access in the Andes vendor extension probing code

 - A new key for scalar misaligned access performance in hwprobe, which
   correctly treat the values as an enum (as opposed to a bitmap)

* tag 'riscv-for-linus-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  riscv: Fix out-of-bounds when accessing Andes per hart vendor extension array
  RISC-V: hwprobe: Add SCALAR to misaligned perf defines
  RISC-V: hwprobe: Add MISALIGNED_PERF key
  RISC-V: ACPI: NUMA: initialize all values of acpi_early_node_map to NUMA_NO_NODE
  riscv: change XIP's kernel_map.size to be size of the entire kernel
  riscv: entry: always initialize regs->a0 to -ENOSYS
  riscv: Re-introduce global icache flush in patch_text_XXX()

Merge tag 'trace-v6.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

Pull tracing fixes from Steven Rostedt:
 "A couple of fixes for tracing:

   - Prevent a NULL pointer dereference in the error path of RTLA tool

   - Fix an infinite loop bug when reading from the ring buffer when
     closed. If there's a thread trying to read the ring buffer and it
     gets closed by another thread, the one reading will go into an
     infinite loop when the buffer is empty instead of exiting back to
     user space"

* tag 'trace-v6.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  rtla/osnoise: Prevent NULL dereference in error handling
  tracing: Return from tracing_buffers_read() if the file has been closed

Merge tag 'keys-trusted-next-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd

Pull key fixes from Jarkko Sakkinen:
 "Two bug fixes for a memory corruption bug and a memory leak bug in the
  DCP trusted keys type.

  Just as a reminder DCP was a crypto coprocessor in i.MX SoCs"

* tag 'keys-trusted-next-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd:
  KEYS: trusted: dcp: fix leak of blob encryption key
  KEYS: trusted: fix DCP blob payload length assignment

bcachefs: fix incorrect i_state usage

Reported-by: [email protected]
Signed-off-by: Kent Overstreet <[email protected]>

bcachefs: avoid overflowing LRU_TIME_BITS for cached data lru

Reported-by: [email protected]
Signed-off-by: Kent Overstreet <[email protected]>

bcachefs: Fix forgetting to pass trans to fsck_err()

Reported-by: [email protected]
Signed-off-by: Kent Overstreet <[email protected]>

bcachefs: Increase size of cuckoo hash table on too many rehashes

Also, improve the calculation of the new table size, so that it can
shrink when needed.

Signed-off-by: Kent Overstreet <[email protected]>

Merge tag 'for-6.11/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mikulas Patocka:

 - fix misbehavior if suspend or resume is interrupted by a signal

 - fix wrong indentation in dm-crypt.rst

 - fix memory allocation failure in dm-persistent-data

* tag 'for-6.11/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm persistent data: fix memory allocation failure
  Documentation: dm-crypt.rst warning + error fix
  dm resume: don't return EINVAL when signalled
  dm suspend: return -ERESTARTSYS instead of -EINTR

Merge tag 'iommu-fixes-v6.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/iommu/linux

Pull iommu fixes from Joerg Roedel:

 - Bring back a lost return statement in io-page-fault code

 - Remove an unused function declaration

* tag 'iommu-fixes-v6.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/iommu/linux:
  iommu: Remove unused declaration iommu_sva_unbind_gpasid()
  iommu: Restore lost return in iommu_report_device_fault()

Merge tag 'gpio-fixes-for-v6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux

Pull gpio fix from Bartosz Golaszewski:

 - add the shutdown() callback to gpio-mlxbf3 in order to disable
   interrupts during graceful reboot

* tag 'gpio-fixes-for-v6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpio: mlxbf3: Support shutdown() function

Merge tag 'sound-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "All small fixes, mostly for usual suspects, HD-audio and USB-audio
  device-specific fixes / quirks. The Cirrus codec support took the
  update of SPI header as well. Other than that, there is a regression
  fix in the sanity check of ALSA timer code"

* tag 'sound-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda/tas2781: Use correct endian conversion
  ALSA: usb-audio: Support Yamaha P-125 quirk entry
  ALSA: hda: cs35l41: Remove redundant call to hda_cs_dsp_control_remove()
  ALSA: hda: cs35l56: Remove redundant call to hda_cs_dsp_control_remove()
  ALSA: hda/tas2781: fix wrong calibrated data order
  ALSA: usb-audio: Add delay quirk for VIVO USB-C-XE710 HEADSET
  ALSA: hda/realtek: Add support for new HP G12 laptops
  ALSA: hda/realtek: Fix noise from speakers on Lenovo IdeaPad 3 15IAU7
  ALSA: timer: Relax start tick time check for slave timer elements
  spi: Add empty versions of ACPI functions

Merge tag 'drm-fixes-2024-08-16' of https://gitlab.freedesktop.org/drm/kernel

Pull drm fixes from Dave Airlie:
 "Weekly drm fixes, mostly amdgpu and xe. The larger amdgpu fix is for a
  new IP block introduced in rc1, so should be fine. The xe fixes
  contain some missed fixes from the end of the previous round along
  with some fixes which required precursor changes, but otherwise
  everything seems fine,

  mediatek:
   - fix cursor crash

  amdgpu:
   - Fix MES ring buffer overflow
   - DCN 3.5 fix
   - DCN 3.2.1 fix
   - DP MST fix
   - Cursor fixes
   - JPEG fixes
   - Context ops validation
   - MES 12 fixes
   - VCN 5.0 fix
   - HDP fix

  panel:
   - dt bindings style fix
   - orientation quirks

  rockchip:
   - inno-hdmi: fix infoframe upload

  v3d:
   - fix OOB access in v3d_csd_job_run()

  xe:
   - Validate user fence during creation
   - Fix use after free when client stats are captured
   - SRIOV fixes
   - Runtime PM fixes"

* tag 'drm-fixes-2024-08-16' of https://gitlab.freedesktop.org/drm/kernel: (37 commits)
  drm/xe: Hold a PM ref when GT TLB invalidations are inflight
  drm/xe: Drop xe_gt_tlb_invalidation_wait
  drm/xe: Add xe_gt_tlb_invalidation_fence_init helper
  drm/xe/pf: Fix VF config validation on multi-GT platforms
  drm/xe: Build PM into GuC CT layer
  drm/xe/vf: Fix register value lookup
  drm/xe: Fix use after free when client stats are captured
  drm/xe: Take a ref to xe file when user creates a VM
  drm/xe: Add ref counting for xe_file
  drm/xe: Move part of xe_file cleanup to a helper
  drm/xe: Validate user fence during creation
  drm/rockchip: inno-hdmi: Fix infoframe upload
  drm/amd/amdgpu: add HDP_SD support on gc 12.0.0/1
  drm/amdgpu: Update kmd_fw_shared for VCN5
  drm/amd/amdgpu: command submission parser for JPEG
  drm/amdgpu/mes12: fix suspend issue
  drm/amdgpu/mes12: sw/hw fini for unified mes
  drm/amdgpu/mes12: configure two pipes hardware resources
  drm/amdgpu/mes12: adjust mes12 sw/hw init for multiple pipes
  drm/amdgpu/mes12: add mes pipe switch support
  ...

Merge tag 'i2c-host-fixes-6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/andi.shyti/linux into i2c/for-current

Two fixes in this update:

Tegra I2C Controller: Addresses a potential double-locking issue
during probe. ACPI devices are not IRQ-safe when invoking runtime
suspend and resume functions, so the irq_safe flag should not be
set.

Qualcomm GENI I2C Controller: Fixes an oversight in the exit path
of the runtime_resume() function, which was missed in the
previous release.

ASoC: SOF: amd: Fix for acp init sequence

When ACP is not powered on by default, acp power on sequence explicitly
invoked by programming pgfsm control mask. The existing implementation
checks the same PGFSM status mask and programs the same PGFSM control mask
in all ACP variants which breaks acp power on sequence for ACP6.0 and
ACP6.3 variants. So to fix this issue, update ACP pgfsm control mask and
status mask based on acp descriptor rev field, which will vary based on
acp variant.

Fixes: 846aef1d7cc0 ("ASoC: SOF: amd: Add Renoir ACP HW support")
Signed-off-by: Vijendar Mukunda <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>

thermal: gov_bang_bang: Use governor_data to reduce overhead

After running once, the for_each_trip_desc() loop in
bang_bang_manage() is pure needless overhead because it is not going to
make any changes unless a new cooling device has been bound to one of
the trips in the thermal zone or the system is resuming from sleep.

For this reason, make bang_bang_manage() set governor_data for the
thermal zone and check it upfront to decide whether or not it needs to
do anything.

However, governor_data needs to be reset in some cases to let
bang_bang_manage() know that it should walk the trips again, so add an
.update_tz() callback to the governor and make the core additionally
invoke it during system resume.

To avoid affecting the other users of that callback unnecessarily, add
a special notification reason for system resume, THERMAL_TZ_RESUME, and
also pass it to __thermal_zone_device_update() called during system
resume for consistency.

Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Peter Kästle <[email protected]>
Reviewed-by: Zhang Rui <[email protected]>
Cc: 6.10+ <[email protected]> # 6.10+
Link: https://patch.msgid.link/[email protected]

thermal: gov_bang_bang: Add .manage() callback

After recent changes, the Bang-bang governor may not adjust the
initial configuration of cooling devices to the actual situation.

Namely, if a cooling device bound to a certain trip point starts in
the "on" state and the thermal zone temperature is below the threshold
of that trip point, the trip point may never be crossed on the way up
in which case the state of the cooling device will never be adjusted
because the thermal core will never invoke the governor's
.trip_crossed() callback.  [Note that there is no issue if the zone
temperature is at the trip threshold or above it to start with because
.trip_crossed() will be invoked then to indicate the start of thermal
mitigation for the given trip.]

To address this, add a .manage() callback to the Bang-bang governor
and use it to ensure that all of the thermal instances managed by the
governor have been initialized properly and the states of all of the
cooling devices involved have been adjusted to the current zone
temperature as appropriate.

Fixes: 530c932bdf75 ("thermal: gov_bang_bang: Use .trip_crossed() instead of .throttle()")
Link: https://lore.kernel.org/linux-pm/[email protected]/
Cc: 6.10+ <[email protected]> # 6.10+
Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Peter Kästle <[email protected]>
Reviewed-by: Zhang Rui <[email protected]>
Link: https://patch.msgid.link/[email protected]

thermal: gov_bang_bang: Split bang_bang_control()

Move the setting of the thermal instance target state from
bang_bang_control() into a separate function that will be also called
in a different place going forward.

No intentional functional impact.

Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Peter Kästle <[email protected]>
Reviewed-by: Zhang Rui <[email protected]>
Cc: 6.10+ <[email protected]> # 6.10+
Link: https://patch.msgid.link/[email protected]

thermal: gov_bang_bang: Call __thermal_cdev_update() directly

Instead of clearing the "updated" flag for each cooling device
affected by the trip point crossing in bang_bang_control() and
walking all thermal instances to run thermal_cdev_update() for all
of the affected cooling devices, call __thermal_cdev_update()
directly for each of them.

No intentional functional impact.

Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Peter Kästle <[email protected]>
Reviewed-by: Zhang Rui <[email protected]>
Cc: 6.10+ <[email protected]> # 6.10+
Link: https://patch.msgid.link/[email protected]

char: xillybus: Check USB endpoints when probing device

Ensure, as the driver probes the device, that all endpoints that the
driver may attempt to access exist and are of the correct type.

All XillyUSB devices must have a Bulk IN and Bulk OUT endpoint at
address 1. This is verified in xillyusb_setup_base_eps().

On top of that, a XillyUSB device may have additional Bulk OUT
endpoints. The information about these endpoints' addresses is deduced
from a data structure (the IDT) that the driver fetches from the device
while probing it. These endpoints are checked in setup_channels().

A XillyUSB device never has more than one IN endpoint, as all data
towards the host is multiplexed in this single Bulk IN endpoint. This is
why setup_channels() only checks OUT endpoints.

Reported-by: [email protected]
Cc: stable <[email protected]>
Closes: https://lore.kernel.org/all/[email protected]/T/
Fixes: a53d1202aef1 ("char: xillybus: Add driver for XillyUSB (Xillybus variant for USB)").
Signed-off-by: Eli Billauer <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

char: xillybus: Refine workqueue handling

As the wakeup work item now runs on a separate workqueue, it needs to be
flushed separately along with flushing the device's workqueue.

Also, move the destroy_workqueue() call to the end of the exit method,
so that deinitialization is done in the opposite order of
initialization.

Fixes: ccbde4b128ef ("char: xillybus: Don't destroy workqueue from work item running on it")
Cc: stable <[email protected]>
Signed-off-by: Eli Billauer <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

mm/migrate: fix deadlock in migrate_pages_batch() on large folios

Currently, migrate_pages_batch() can lock multiple locked folios with an
arbitrary order.  Although folio_trylock() is used to avoid deadlock as
commit 2ef7dbb26990 ("migrate_pages: try migrate in batch asynchronously
firstly") mentioned, it seems try_split_folio() is still missing.

It was found by compaction stress test when I explicitly enable EROFS
compressed files to use large folios, which case I cannot reproduce with
the same workload if large folio support is off (current mainline).
Typically, filesystem reads (with locked file-backed folios) could use
another bdev/meta inode to load some other I/Os (e.g.  inode extent
metadata or caching compressed data), so the locking order will be:

  file-backed folios  (A)
     bdev/meta folios (B)

The following calltrace shows the deadlock:
   Thread 1 takes (B) lock and tries to take folio (A) lock
   Thread 2 takes (A) lock and tries to take folio (B) lock

[Thread 1]
INFO: task stress:1824 blocked for more than 30 seconds.
      Tainted: G           OE      6.10.0-rc7+ #6
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:stress          state:D stack:0     pid:1824  tgid:1824  ppid:1822   flags:0x0000000c
Call trace:
 __switch_to+0xec/0x138
 __schedule+0x43c/0xcb0
 schedule+0x54/0x198
 io_schedule+0x44/0x70
 folio_wait_bit_common+0x184/0x3f8
<-- folio mapping ffff00036d69cb18 index 996  (**)
 __folio_lock+0x24/0x38
 migrate_pages_batch+0x77c/0xea0 // try_split_folio (mm/migrate.c:1486:2)
// migrate_pages_batch (mm/migrate.c:1734:16)
<--- LIST_HEAD(unmap_folios) has
..
folio mapping 0xffff0000d184f1d8 index 1711;   (*)
folio mapping 0xffff0000d184f1d8 index 1712;
..
 migrate_pages+0xb28/0xe90
 compact_zone+0xa08/0x10f0
 compact_node+0x9c/0x180
 sysctl_compaction_handler+0x8c/0x118
 proc_sys_call_handler+0x1a8/0x280
 proc_sys_write+0x1c/0x30
 vfs_write+0x240/0x380
 ksys_write+0x78/0x118
 __arm64_sys_write+0x24/0x38
 invoke_syscall+0x78/0x108
 el0_svc_common.constprop.0+0x48/0xf0
 do_el0_svc+0x24/0x38
 el0_svc+0x3c/0x148
 el0t_64_sync_handler+0x100/0x130
 el0t_64_sync+0x190/0x198

[Thread 2]
INFO: task stress:1825 blocked for more than 30 seconds.
      Tainted: G           OE      6.10.0-rc7+ #6
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:stress          state:D stack:0     pid:1825  tgid:1825  ppid:1822   flags:0x0000000c
Call trace:
 __switch_to+0xec/0x138
 __schedule+0x43c/0xcb0
 schedule+0x54/0x198
 io_schedule+0x44/0x70
 folio_wait_bit_common+0x184/0x3f8
<-- folio = 0xfffffdffc6b503c0 (mapping == 0xffff0000d184f1d8 index == 1711) (*)
 __folio_lock+0x24/0x38
 z_erofs_runqueue+0x384/0x9c0 [erofs]
 z_erofs_readahead+0x21c/0x350 [erofs]       <-- folio mapping 0xffff00036d69cb18 range from [992, 1024] (**)
 read_pages+0x74/0x328
 page_cache_ra_order+0x26c/0x348
 ondemand_readahead+0x1c0/0x3a0
 page_cache_sync_ra+0x9c/0xc0
 filemap_get_pages+0xc4/0x708
 filemap_read+0x104/0x3a8
 generic_file_read_iter+0x4c/0x150
 vfs_read+0x27c/0x330
 ksys_pread64+0x84/0xd0
 __arm64_sys_pread64+0x28/0x40
 invoke_syscall+0x78/0x108
 el0_svc_common.constprop.0+0x48/0xf0
 do_el0_svc+0x24/0x38
 el0_svc+0x3c/0x148
 el0t_64_sync_handler+0x100/0x130
 el0t_64_sync+0x190/0x198

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 5dfab109d519 ("migrate_pages: batch _unmap and _move")
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: "Huang, Ying" <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

alloc_tag: mark pages reserved during CMA activation as not tagged

During CMA activation, pages in CMA area are prepared and then freed
without being allocated.  This triggers warnings when memory allocation
debug config (CONFIG_MEM_ALLOC_PROFILING_DEBUG) is enabled.  Fix this by
marking these pages not tagged before freeing them.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty")
Signed-off-by: Suren Baghdasaryan <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: Kent Overstreet <[email protected]>
Cc: Pasha Tatashin <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: <[email protected]> [6.10]
Signed-off-by: Andrew Morton <[email protected]>

alloc_tag: introduce clear_page_tag_ref() helper function

In several cases we are freeing pages which were not allocated using
common page allocators.  For such cases, in order to keep allocation
accounting correct, we should clear the page tag to indicate that the page
being freed is expected to not have a valid allocation tag.  Introduce
clear_page_tag_ref() helper function to be used for this.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty")
Signed-off-by: Suren Baghdasaryan <[email protected]>
Suggested-by: David Hildenbrand <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Reviewed-by: Pasha Tatashin <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: Kent Overstreet <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: <[email protected]> [6.10]
Signed-off-by: Andrew Morton <[email protected]>

crash: fix riscv64 crash memory reserve dead loop

On RISCV64 Qemu machine with 512MB memory, cmdline "crashkernel=500M,high"
will cause system stall as below:

 Zone ranges:
   DMA32    [mem 0x0000000080000000-0x000000009fffffff]
   Normal   empty
 Movable zone start for each node
 Early memory node ranges
   node   0: [mem 0x0000000080000000-0x000000008005ffff]
   node   0: [mem 0x0000000080060000-0x000000009fffffff]
 Initmem setup node 0 [mem 0x0000000080000000-0x000000009fffffff]
(stall here)

commit 5d99cadf1568 ("crash: fix x86_32 crash memory reserve dead loop
bug") fix this on 32-bit architecture.  However, the problem is not
completely solved.  If `CRASH_ADDR_LOW_MAX = CRASH_ADDR_HIGH_MAX` on
64-bit architecture, for example, when system memory is equal to
CRASH_ADDR_LOW_MAX on RISCV64, the following infinite loop will also
occur:

-> reserve_crashkernel_generic() and high is true
   -> alloc at [CRASH_ADDR_LOW_MAX, CRASH_ADDR_HIGH_MAX] fail
      -> alloc at [0, CRASH_ADDR_LOW_MAX] fail and repeatedly
         (because CRASH_ADDR_LOW_MAX = CRASH_ADDR_HIGH_MAX).

As Catalin suggested, do not remove the ",high" reservation fallback to
",low" logic which will change arm64's kdump behavior, but fix it by
skipping the above situation similar to commit d2f32f23190b ("crash: fix
x86_32 crash memory reserve dead loop").

After this patch, it print:
cannot allocate crashkernel (size:0x1f400000)

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Jinjie Ruan <[email protected]>
Suggested-by: Catalin Marinas <[email protected]>
Reviewed-by: Catalin Marinas <[email protected]>
Acked-by: Baoquan He <[email protected]>
Cc: Albert Ou <[email protected]>
Cc: Dave Young <[email protected]>
Cc: Palmer Dabbelt <[email protected]>
Cc: Paul Walmsley <[email protected]>
Cc: Vivek Goyal <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

selftests: memfd_secret: don't build memfd_secret test on unsupported arches

[1] mentions that memfd_secret is only supported on arm64, riscv, x86 and
x86_64 for now.  It doesn't support other architectures.  I found the
build error on arm and decided to send the fix as it was creating noise on
KernelCI:

memfd_secret.c: In function 'memfd_secret':
memfd_secret.c:42:24: error: '__NR_memfd_secret' undeclared (first use in this function);
did you mean 'memfd_secret'?
   42 |         return syscall(__NR_memfd_secret, flags);
      |                        ^~~~~~~~~~~~~~~~~
      |                        memfd_secret

Hence I'm adding condition that memfd_secret should only be compiled on
supported architectures.

Also check in run_vmtests script if memfd_secret binary is present before
executing it.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lore.kernel.org/all/[email protected]/
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 76fe17ef588a ("secretmem: test: add basic selftest for memfd_secret(2)")
Signed-off-by: Muhammad Usama Anjum <[email protected]>
Reviewed-by: Shuah Khan <[email protected]>
Acked-by: Mike Rapoport (Microsoft) <[email protected]>
Cc: Albert Ou <[email protected]>
Cc: James Bottomley <[email protected]>
Cc: Mike Rapoport (Microsoft) <[email protected]>
Cc: Palmer Dabbelt <[email protected]>
Cc: Paul Walmsley <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm: fix endless reclaim on machines with unaccepted memory

Unaccepted memory is considered unusable free memory, which is not counted
as free on the zone watermark check.  This causes get_page_from_freelist()
to accept more memory to hit the high watermark, but it creates problems
in the reclaim path.

The reclaim path encounters a failed zone watermark check and attempts to
reclaim memory.  This is usually successful, but if there is little or no
reclaimable memory, it can result in endless reclaim with little to no
progress.  This can occur early in the boot process, just after start of
the init process when the only reclaimable memory is the page cache of the
init executable and its libraries.

Make unaccepted memory free from watermark check point of view.  This way
unaccepted memory will never be the trigger of memory reclaim.  Accept
more memory in the get_page_from_freelist() if needed.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: dcdfdd40fa82 ("mm: Add support for unaccepted memory")
Signed-off-by: Kirill A. Shutemov <[email protected]>
Reported-by: Jianxiong Gao <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Tested-by: Jianxiong Gao <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Kirill A. Shutemov <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Mike Rapoport (Microsoft) <[email protected]>
Cc: Tom Lendacky <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: <[email protected]> [6.5+]
Signed-off-by: Andrew Morton <[email protected]>

selftests/mm: compaction_test: fix off by one in check_compaction()

The "initial_nr_hugepages" variable is unsigned long so it takes up to 20
characters to print, plus 1 more character for the NUL terminator.
Unfortunately, this buffer is not quite large enough for the terminator to
fit.  Also use snprintf() for a belt and suspenders approach.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: fb9293b6b015 ("selftests/mm: compaction_test: fix bogus test success and reduce probability of OOM-killer invocation")
Signed-off-by: Dan Carpenter <[email protected]>
Cc: Shuah Khan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm/numa: no task_numa_fault() call if PMD is changed

When handling a numa page fault, task_numa_fault() should be called by a
process that restores the page table of the faulted folio to avoid
duplicated stats counting.  Commit c5b5a3dd2c1f ("mm: thp: refactor NUMA
fault handling") restructured do_huge_pmd_numa_page() and did not avoid
task_numa_fault() call in the second page table check after a numa
migration failure.  Fix it by making all !pmd_same() return immediately.

This issue can cause task_numa_fault() being called more than necessary
and lead to unexpected numa balancing results (It is hard to tell whether
the issue will cause positive or negative performance impact due to
duplicated numa fault counting).

Link: https://lkml.kernel.org/r/[email protected]
Fixes: c5b5a3dd2c1f ("mm: thp: refactor NUMA fault handling")
Reported-by: "Huang, Ying" <[email protected]>
Closes: https://lore.kernel.org/linux-mm/[email protected]/
Signed-off-by: Zi Yan <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Cc: Baolin Wang <[email protected]>
Cc: "Huang, Ying" <[email protected]>
Cc: Kefeng Wang <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Yang Shi <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm/numa: no task_numa_fault() call if PTE is changed

When handling a numa page fault, task_numa_fault() should be called by a
process that restores the page table of the faulted folio to avoid
duplicated stats counting.  Commit b99a342d4f11 ("NUMA balancing: reduce
TLB flush via delaying mapping on hint page fault") restructured
do_numa_page() and did not avoid task_numa_fault() call in the second page
table check after a numa migration failure.  Fix it by making all
!pte_same() return immediately.

This issue can cause task_numa_fault() being called more than necessary
and lead to unexpected numa balancing results (It is hard to tell whether
the issue will cause positive or negative performance impact due to
duplicated numa fault counting).

Link: https://lkml.kernel.org/r/[email protected]
Fixes: b99a342d4f11 ("NUMA balancing: reduce TLB flush via delaying mapping on hint page fault")
Signed-off-by: Zi Yan <[email protected]>
Reported-by: "Huang, Ying" <[email protected]>
Closes: https://lore.kernel.org/linux-mm/[email protected]/
Acked-by: David Hildenbrand <[email protected]>
Cc: Baolin Wang <[email protected]>
Cc: Kefeng Wang <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Yang Shi <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm/vmalloc: fix page mapping if vm_area_alloc_pages() with high order fallback to order 0

The __vmap_pages_range_noflush() assumes its argument pages** contains
pages with the same page shift.  However, since commit e9c3cda4d86e ("mm,
vmalloc: fix high order __GFP_NOFAIL allocations"), if gfp_flags includes
__GFP_NOFAIL with high order in vm_area_alloc_pages() and page allocation
failed for high order, the pages** may contain two different page shifts
(high order and order-0).  This could lead __vmap_pages_range_noflush() to
perform incorrect mappings, potentially resulting in memory corruption.

Users might encounter this as follows (vmap_allow_huge = true, 2M is for
PMD_SIZE):

kvmalloc(2M, __GFP_NOFAIL|GFP_X)
    __vmalloc_node_range_noprof(vm_flags=VM_ALLOW_HUGE_VMAP)
        vm_area_alloc_pages(order=9) ---> order-9 allocation failed and fallback to order-0
            vmap_pages_range()
                vmap_pages_range_noflush()
                    __vmap_pages_range_noflush(page_shift = 21) ----> wrong mapping happens

We can remove the fallback code because if a high-order allocation fails,
__vmalloc_node_range_noprof() will retry with order-0.  Therefore, it is
unnecessary to fallback to order-0 here.  Therefore, fix this by removing
the fallback code.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: e9c3cda4d86e ("mm, vmalloc: fix high order __GFP_NOFAIL allocations")
Signed-off-by: Hailong Liu <[email protected]>
Reported-by: Tangquan Zheng <[email protected]>
Reviewed-by: Baoquan He <[email protected]>
Reviewed-by: Uladzislau Rezki (Sony) <[email protected]>
Acked-by: Barry Song <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm/memory-failure: use raw_spinlock_t in struct memory_failure_cpu

The memory_failure_cpu structure is a per-cpu structure.  Access to its
content requires the use of get_cpu_var() to lock in the current CPU and
disable preemption.  The use of a regular spinlock_t for locking purpose
is fine for a non-RT kernel.

Since the integration of RT spinlock support into the v5.15 kernel, a
spinlock_t in a RT kernel becomes a sleeping lock and taking a sleeping
lock in a preemption disabled context is illegal resulting in the
following kind of warning.

  [12135.732244] BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48
  [12135.732248] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 270076, name: kworker/0:0
  [12135.732252] preempt_count: 1, expected: 0
  [12135.732255] RCU nest depth: 2, expected: 2
    :
  [12135.732420] Hardware name: Dell Inc. PowerEdge R640/0HG0J8, BIOS 2.10.2 02/24/2021
  [12135.732423] Workqueue: kacpi_notify acpi_os_execute_deferred
  [12135.732433] Call Trace:
  [12135.732436]  <TASK>
  [12135.732450]  dump_stack_lvl+0x57/0x81
  [12135.732461]  __might_resched.cold+0xf4/0x12f
  [12135.732479]  rt_spin_lock+0x4c/0x100
  [12135.732491]  memory_failure_queue+0x40/0xe0
  [12135.732503]  ghes_do_memory_failure+0x53/0x390
  [12135.732516]  ghes_do_proc.constprop.0+0x229/0x3e0
  [12135.732575]  ghes_proc+0xf9/0x1a0
  [12135.732591]  ghes_notify_hed+0x6a/0x150
  [12135.732602]  notifier_call_chain+0x43/0xb0
  [12135.732626]  blocking_notifier_call_chain+0x43/0x60
  [12135.732637]  acpi_ev_notify_dispatch+0x47/0x70
  [12135.732648]  acpi_os_execute_deferred+0x13/0x20
  [12135.732654]  process_one_work+0x41f/0x500
  [12135.732695]  worker_thread+0x192/0x360
  [12135.732715]  kthread+0x111/0x140
  [12135.732733]  ret_from_fork+0x29/0x50
  [12135.732779]  </TASK>

Fix it by using a raw_spinlock_t for locking instead.

Also move the pr_err() out of the lock critical section and after
put_cpu_ptr() to avoid indeterminate latency and the possibility of sleep
with this call.

[[email protected]: don't hold percpu ref across pr_err(), per Miaohe]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 0f383b6dc96e ("locking/spinlock: Provide RT variant")
Signed-off-by: Waiman Long <[email protected]>
Acked-by: Miaohe Lin <[email protected]>
Cc: "Huang, Ying" <[email protected]>
Cc: Juri Lelli <[email protected]>
Cc: Len Brown <[email protected]>
Cc: Naoya Horiguchi <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm: don't account memmap per-node

Fix invalid access to pgdat during hot-remove operation:
ndctl users reported a GPF when trying to destroy a namespace:
$ ndctl destroy-namespace all -r all -f
 Segmentation fault
 dmesg:
 Oops: general protection fault, probably for
 non-canonical address 0xdffffc0000005650: 0000 [#1] PREEMPT SMP KASAN
 PTI
 KASAN: probably user-memory-access in range
 [0x000000000002b280-0x000000000002b287]
 CPU: 26 UID: 0 PID: 1868 Comm: ndctl Not tainted 6.11.0-rc1 #1
 Hardware name: Dell Inc. PowerEdge R640/08HT8T, BIOS
 2.20.1 09/13/2023
 RIP: 0010:mod_node_page_state+0x2a/0x110

cxl-test users report a GPF when trying to unload the test module:
$ modrpobe -r cxl-test
 dmesg
 BUG: unable to handle page fault for address: 0000000000004200
 #PF: supervisor read access in kernel mode
 #PF: error_code(0x0000) - not-present page
 PGD 0 P4D 0
 Oops: Oops: 0000 [#1] PREEMPT SMP PTI
 CPU: 0 UID: 0 PID: 1076 Comm: modprobe Tainted: G O N 6.11.0-rc1 #197
 Tainted: [O]=OOT_MODULE, [N]=TEST
 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/15
 RIP: 0010:mod_node_page_state+0x6/0x90

Currently, when memory is hot-plugged or hot-removed the accounting is
done based on the assumption that memmap is allocated from the same node
as the hot-plugged/hot-removed memory, which is not always the case.

In addition, there are challenges with keeping the node id of the memory
that is being remove to the time when memmap accounting is actually
performed: since this is done after remove_pfn_range_from_zone(), and
also after remove_memory_block_devices(). Meaning that we cannot use
pgdat nor walking though memblocks to get the nid.

Given all of that, account the memmap overhead system wide instead.

For this we are going to be using global atomic counters, but given that
memmap size is rarely modified, and normally is only modified either
during early boot when there is only one CPU, or under a hotplug global
mutex lock, therefore there is no need for per-cpu optimizations.

Also, while we are here rename nr_memmap to nr_memmap_pages, and
nr_memmap_boot to nr_memmap_boot_pages to be self explanatory that the
units are in page count.

[[email protected]: address a few nits from David Hildenbrand]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 15995a352474 ("mm: report per-page metadata information")
Signed-off-by: Pasha Tatashin <[email protected]>
Reported-by: Yi Zhang <[email protected]>
Closes: https://lore.kernel.org/linux-cxl/CAHj4cs9Ax1=CoJkgBGP_+sNu6-6=6v=_L-ZBZY0bVLD3wUWZQg@mail.gmail.com
Reported-by: Alison Schofield <[email protected]>
Closes: https://lore.kernel.org/linux-mm/Zq0tPd2h6alFz8XF@aschofie-mobl2/#t
Tested-by: Dan Williams <[email protected]>
Tested-by: Alison Schofield <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Acked-by: David Rientjes <[email protected]>
Tested-by: Yi Zhang <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: Fan Ni <[email protected]>
Cc: Joel Granados <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Li Zhijian <[email protected]>
Cc: Matthew Wilcox (Oracle) <[email protected]>
Cc: Mike Rapoport <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: Nhat Pham <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Yosry Ahmed <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm: add system wide stats items category

/proc/vmstat contains events and stats, events can only grow, but stats
can grow and shrink.

vmstat has the following:
-------------------------
NR_VM_ZONE_STAT_ITEMS: per-zone stats
NR_VM_NUMA_EVENT_ITEMS: per-numa events
NR_VM_NODE_STAT_ITEMS: per-numa stats
NR_VM_WRITEBACK_STAT_ITEMS: system-wide background-writeback and
dirty-throttling tresholds.
NR_VM_EVENT_ITEMS: system-wide events
-------------------------

Rename NR_VM_WRITEBACK_STAT_ITEMS to NR_VM_STAT_ITEMS, to track the
system-wide stats, we are going to add per-page metadata stats to this
category in the next patch.

Also delete unused writeback_stat_name().

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 15995a352474 ("mm: report per-page metadata information")
Signed-off-by: Pasha Tatashin <[email protected]>
Suggested-by: Yosry Ahmed <[email protected]>
Tested-by: Alison Schofield <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Acked-by: David Rientjes <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: Joel Granados <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Li Zhijian <[email protected]>
Cc: Matthew Wilcox (Oracle) <[email protected]>
Cc: Mike Rapoport <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: Nhat Pham <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Yi Zhang <[email protected]>
Cc: Fan Ni <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm: don't account memmap on failure

Patch series "Fixes for memmap accounting", v4.

Memmap accounting provides us with observability of how much memory is
used for per-page metadata: i.e. "struct page"'s and "struct page_ext".
It also provides with information of how much was allocated using
boot allocator (i.e. not part of MemTotal), and how much was allocated
using buddy allocated (i.e. part of MemTotal).

This small series fixes a few problems that were discovered with the
original patch.

This patch (of 3):

When we fail to allocate the mmemmap in alloc_vmemmap_page_list(), do not
account any already-allocated pages: we're going to free all them before
we return from the function.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 15995a352474 ("mm: report per-page metadata information")
Signed-off-by: Pasha Tatashin <[email protected]>
Reviewed-by: Fan Ni <[email protected]>
Reviewed-by: Yosry Ahmed <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Tested-by: Alison Schofield <[email protected]>
Reviewed-by: Muchun Song <[email protected]>
Acked-by: David Rientjes <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: Joel Granados <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Li Zhijian <[email protected]>
Cc: Matthew Wilcox (Oracle) <[email protected]>
Cc: Mike Rapoport <[email protected]>
Cc: Nhat Pham <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Yi Zhang <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mm/hugetlb: fix hugetlb vs. core-mm PT locking

We recently made GUP's common page table walking code to also walk hugetlb
VMAs without most hugetlb special-casing, preparing for the future of
having less hugetlb-specific page table walking code in the codebase.
Turns out that we missed one page table locking detail: page table locking
for hugetlb folios that are not mapped using a single PMD/PUD.

Assume we have hugetlb folio that spans multiple PTEs (e.g., 64 KiB
hugetlb folios on arm64 with 4 KiB base page size).  GUP, as it walks the
page tables, will perform a pte_offset_map_lock() to grab the PTE table
lock.

However, hugetlb that concurrently modifies these page tables would
actually grab the mm->page_table_lock: with USE_SPLIT_PTE_PTLOCKS, the
locks would differ.  Something similar can happen right now with hugetlb
folios that span multiple PMDs when USE_SPLIT_PMD_PTLOCKS.

This issue can be reproduced [1], for example triggering:

[ 3105.936100] ------------[ cut here ]------------
[ 3105.939323] WARNING: CPU: 31 PID: 2732 at mm/gup.c:142 try_grab_folio+0x11c/0x188
[ 3105.944634] Modules linked in: [...]
[ 3105.974841] CPU: 31 PID: 2732 Comm: reproducer Not tainted 6.10.0-64.eln141.aarch64 #1
[ 3105.980406] Hardware name: QEMU KVM Virtual Machine, BIOS edk2-20240524-4.fc40 05/24/2024
[ 3105.986185] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 3105.991108] pc : try_grab_folio+0x11c/0x188
[ 3105.994013] lr : follow_page_pte+0xd8/0x430
[ 3105.996986] sp : ffff80008eafb8f0
[ 3105.999346] x29: ffff80008eafb900 x28: ffffffe8d481f380 x27: 00f80001207cff43
[ 3106.004414] x26: 0000000000000001 x25: 0000000000000000 x24: ffff80008eafba48
[ 3106.009520] x23: 0000ffff9372f000 x22: ffff7a54459e2000 x21: ffff7a546c1aa978
[ 3106.014529] x20: ffffffe8d481f3c0 x19: 0000000000610041 x18: 0000000000000001
[ 3106.019506] x17: 0000000000000001 x16: ffffffffffffffff x15: 0000000000000000
[ 3106.024494] x14: ffffb85477fdfe08 x13: 0000ffff9372ffff x12: 0000000000000000
[ 3106.029469] x11: 1fffef4a88a96be1 x10: ffff7a54454b5f0c x9 : ffffb854771b12f0
[ 3106.034324] x8 : 0008000000000000 x7 : ffff7a546c1aa980 x6 : 0008000000000080
[ 3106.038902] x5 : 00000000001207cf x4 : 0000ffff9372f000 x3 : ffffffe8d481f000
[ 3106.043420] x2 : 0000000000610041 x1 : 0000000000000001 x0 : 0000000000000000
[ 3106.047957] Call trace:
[ 3106.049522]  try_grab_folio+0x11c/0x188
[ 3106.051996]  follow_pmd_mask.constprop.0.isra.0+0x150/0x2e0
[ 3106.055527]  follow_page_mask+0x1a0/0x2b8
[ 3106.058118]  __get_user_pages+0xf0/0x348
[ 3106.060647]  faultin_page_range+0xb0/0x360
[ 3106.063651]  do_madvise+0x340/0x598

Let's make huge_pte_lockptr() effectively use the same PT locks as any
core-mm page table walker would.  Add ptep_lockptr() to obtain the PTE
page table lock using a pte pointer -- unfortunately we cannot convert
pte_lockptr() because virt_to_page() doesn't work with kmap'ed page tables
we can have with CONFIG_HIGHPTE.

Handle CONFIG_PGTABLE_LEVELS correctly by checking in reverse order, such
that when e.g., CONFIG_PGTABLE_LEVELS==2 with
PGDIR_SIZE==P4D_SIZE==PUD_SIZE==PMD_SIZE will work as expected.  Document
why that works.

There is one ugly case: powerpc 8xx, whereby we have an 8 MiB hugetlb
folio being mapped using two PTE page tables.  While hugetlb wants to take
the PMD table lock, core-mm would grab the PTE table lock of one of both
PTE page tables.  In such corner cases, we have to make sure that both
locks match, which is (fortunately!) currently guaranteed for 8xx as it
does not support SMP and consequently doesn't use split PT locks.

[1] https://lore.kernel.org/all/1bbfcc7f-f222-45a5-ac44-c5a1381c596d@redhat.com/

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 9cb28da54643 ("mm/gup: handle hugetlb in the generic follow_page_mask code")
Signed-off-by: David Hildenbrand <[email protected]>
Acked-by: Peter Xu <[email protected]>
Reviewed-by: Baolin Wang <[email protected]>
Tested-by: Baolin Wang <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

mseal: fix is_madv_discard()

is_madv_discard did its check wrong. MADV_ flags are not bitwise,
they're normal sequential numbers. So, for instance:
behavior & (/* ... */ | MADV_REMOVE)

tagged both MADV_REMOVE and MADV_RANDOM (bit 0 set) as discard
operations.

As a result the kernel could erroneously block certain madvises (e.g
MADV_RANDOM or MADV_HUGEPAGE) on sealed VMAs due to them sharing bits
with blocked MADV operations (e.g REMOVE or WIPEONFORK).

This is obviously incorrect, so use a switch statement instead.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 8be7258aad44 ("mseal: add mseal syscall")
Signed-off-by: Pedro Falcato <[email protected]>
Tested-by: Jeff Xu <[email protected]>
Reviewed-by: Jeff Xu <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: Liam R. Howlett <[email protected]>
Cc: Shuah Khan <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

Merge tag 'mediatek-drm-fixes-20240805' of https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux into drm-fixes

Mediatek DRM Fixes - 20240805

1. Set sensible cursor width/height values to fix crash

Signed-off-by: Dave Airlie <[email protected]>
From: Chun-Kuang Hu <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

Merge tag 'drm-xe-fixes-2024-08-15' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes

- Validate user fence during creation (Brost)
- Fix use after free when client stats are captured (Umesh)
- SRIOV fixes (Michal)
- Runtime PM fixes (Brost)

Signed-off-by: Dave Airlie <[email protected]>
From: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

Merge tag 'drm-misc-fixes-2024-08-15' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes

Short summary of fixes pull:

panel:
- dt-bindings style fixes

panel-orientation:
- add quirk for Any Loki Max
- add quirk for Any Loki Zero

rockchip:
- inno-hdmi: fix infoframe upload

v3d:
- fix OOB access in v3d_csd_job_run()

Signed-off-by: Dave Airlie <[email protected]>
From: Thomas Zimmermann <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

block: Fix lockdep warning in blk_mq_mark_tag_wait

Lockdep reported a warning in Linux version 6.6:

[  414.344659] ================================
[  414.345155] WARNING: inconsistent lock state
[  414.345658] 6.6.0-07439-gba2303cacfda #6 Not tainted
[  414.346221] --------------------------------
[  414.346712] inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.
[  414.347545] kworker/u10:3/1152 [HC0[0]:SC0[0]:HE0:SE1] takes:
[  414.349245] ffff88810edd1098 (&sbq->ws[i].wait){+.?.}-{2:2}, at: blk_mq_dispatch_rq_list+0x131c/0x1ee0
[  414.351204] {IN-SOFTIRQ-W} state was registered at:
[  414.351751]   lock_acquire+0x18d/0x460
[  414.352218]   _raw_spin_lock_irqsave+0x39/0x60
[  414.352769]   __wake_up_common_lock+0x22/0x60
[  414.353289]   sbitmap_queue_wake_up+0x375/0x4f0
[  414.353829]   sbitmap_queue_clear+0xdd/0x270
[  414.354338]   blk_mq_put_tag+0xdf/0x170
[  414.354807]   __blk_mq_free_request+0x381/0x4d0
[  414.355335]   blk_mq_free_request+0x28b/0x3e0
[  414.355847]   __blk_mq_end_request+0x242/0xc30
[  414.356367]   scsi_end_request+0x2c1/0x830
[  414.345155] WARNING: inconsistent lock state
[  414.345658] 6.6.0-07439-gba2303cacfda #6 Not tainted
[  414.346221] --------------------------------
[  414.346712] inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.
[  414.347545] kworker/u10:3/1152 [HC0[0]:SC0[0]:HE0:SE1] takes:
[  414.349245] ffff88810edd1098 (&sbq->ws[i].wait){+.?.}-{2:2}, at: blk_mq_dispatch_rq_list+0x131c/0x1ee0
[  414.351204] {IN-SOFTIRQ-W} state was registered at:
[  414.351751]   lock_acquire+0x18d/0x460
[  414.352218]   _raw_spin_lock_irqsave+0x39/0x60
[  414.352769]   __wake_up_common_lock+0x22/0x60
[  414.353289]   sbitmap_queue_wake_up+0x375/0x4f0
[  414.353829]   sbitmap_queue_clear+0xdd/0x270
[  414.354338]   blk_mq_put_tag+0xdf/0x170
[  414.354807]   __blk_mq_free_request+0x381/0x4d0
[  414.355335]   blk_mq_free_request+0x28b/0x3e0
[  414.355847]   __blk_mq_end_request+0x242/0xc30
[  414.356367]   scsi_end_request+0x2c1/0x830
[  414.356863]   scsi_io_completion+0x177/0x1610
[  414.357379]   scsi_complete+0x12f/0x260
[  414.357856]   blk_complete_reqs+0xba/0xf0
[  414.358338]   __do_softirq+0x1b0/0x7a2
[  414.358796]   irq_exit_rcu+0x14b/0x1a0
[  414.359262]   sysvec_call_function_single+0xaf/0xc0
[  414.359828]   asm_sysvec_call_function_single+0x1a/0x20
[  414.360426]   default_idle+0x1e/0x30
[  414.360873]   default_idle_call+0x9b/0x1f0
[  414.361390]   do_idle+0x2d2/0x3e0
[  414.361819]   cpu_startup_entry+0x55/0x60
[  414.362314]   start_secondary+0x235/0x2b0
[  414.362809]   secondary_startup_64_no_verify+0x18f/0x19b
[  414.363413] irq event stamp: 428794
[  414.363825] hardirqs last  enabled at (428793): [<ffffffff816bfd1c>] ktime_get+0x1dc/0x200
[  414.364694] hardirqs last disabled at (428794): [<ffffffff85470177>] _raw_spin_lock_irq+0x47/0x50
[  414.365629] softirqs last  enabled at (428444): [<ffffffff85474780>] __do_softirq+0x540/0x7a2
[  414.366522] softirqs last disabled at (428419): [<ffffffff813f65ab>] irq_exit_rcu+0x14b/0x1a0
[  414.367425]
               other info that might help us debug this:
[  414.368194]  Possible unsafe locking scenario:
[  414.368900]        CPU0
[  414.369225]        ----
[  414.369548]   lock(&sbq->ws[i].wait);
[  414.370000]   <Interrupt>
[  414.370342]     lock(&sbq->ws[i].wait);
[  414.370802]
                *** DEADLOCK ***
[  414.371569] 5 locks held by kworker/u10:3/1152:
[  414.372088]  #0: ffff88810130e938 ((wq_completion)writeback){+.+.}-{0:0}, at: process_scheduled_works+0x357/0x13f0
[  414.373180]  #1: ffff88810201fdb8 ((work_completion)(&(&wb->dwork)->work)){+.+.}-{0:0}, at: process_scheduled_works+0x3a3/0x13f0
[  414.374384]  #2: ffffffff86ffbdc0 (rcu_read_lock){....}-{1:2}, at: blk_mq_run_hw_queue+0x637/0xa00
[  414.375342]  #3: ffff88810edd1098 (&sbq->ws[i].wait){+.?.}-{2:2}, at: blk_mq_dispatch_rq_list+0x131c/0x1ee0
[  414.376377]  #4: ffff888106205a08 (&hctx->dispatch_wait_lock){+.-.}-{2:2}, at: blk_mq_dispatch_rq_list+0x1337/0x1ee0
[  414.378607]
               stack backtrace:
[  414.379177] CPU: 0 PID: 1152 Comm: kworker/u10:3 Not tainted 6.6.0-07439-gba2303cacfda #6
[  414.380032] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
[  414.381177] Workqueue: writeback wb_workfn (flush-253:0)
[  414.381805] Call Trace:
[  414.382136]  <TASK>
[  414.382429]  dump_stack_lvl+0x91/0xf0
[  414.382884]  mark_lock_irq+0xb3b/0x1260
[  414.383367]  ? __pfx_mark_lock_irq+0x10/0x10
[  414.383889]  ? stack_trace_save+0x8e/0xc0
[  414.384373]  ? __pfx_stack_trace_save+0x10/0x10
[  414.384903]  ? graph_lock+0xcf/0x410
[  414.385350]  ? save_trace+0x3d/0xc70
[  414.385808]  mark_lock.part.20+0x56d/0xa90
[  414.386317]  mark_held_locks+0xb0/0x110
[  414.386791]  ? __pfx_do_raw_spin_lock+0x10/0x10
[  414.387320]  lockdep_hardirqs_on_prepare+0x297/0x3f0
[  414.387901]  ? _raw_spin_unlock_irq+0x28/0x50
[  414.388422]  trace_hardirqs_on+0x58/0x100
[  414.388917]  _raw_spin_unlock_irq+0x28/0x50
[  414.389422]  __blk_mq_tag_busy+0x1d6/0x2a0
[  414.389920]  __blk_mq_get_driver_tag+0x761/0x9f0
[  414.390899]  blk_mq_dispatch_rq_list+0x1780/0x1ee0
[  414.391473]  ? __pfx_blk_mq_dispatch_rq_list+0x10/0x10
[  414.392070]  ? sbitmap_get+0x2b8/0x450
[  414.392533]  ? __blk_mq_get_driver_tag+0x210/0x9f0
[  414.393095]  __blk_mq_sched_dispatch_requests+0xd99/0x1690
[  414.393730]  ? elv_attempt_insert_merge+0x1b1/0x420
[  414.394302]  ? __pfx___blk_mq_sched_dispatch_requests+0x10/0x10
[  414.394970]  ? lock_acquire+0x18d/0x460
[  414.395456]  ? blk_mq_run_hw_queue+0x637/0xa00
[  414.395986]  ? __pfx_lock_acquire+0x10/0x10
[  414.396499]  blk_mq_sched_dispatch_requests+0x109/0x190
[  414.397100]  blk_mq_run_hw_queue+0x66e/0xa00
[  414.397616]  blk_mq_flush_plug_list.part.17+0x614/0x2030
[  414.398244]  ? __pfx_blk_mq_flush_plug_list.part.17+0x10/0x10
[  414.398897]  ? writeback_sb_inodes+0x241/0xcc0
[  414.399429]  blk_mq_flush_plug_list+0x65/0x80
[  414.399957]  __blk_flush_plug+0x2f1/0x530
[  414.400458]  ? __pfx___blk_flush_plug+0x10/0x10
[  414.400999]  blk_finish_plug+0x59/0xa0
[  414.401467]  wb_writeback+0x7cc/0x920
[  414.401935]  ? __pfx_wb_writeback+0x10/0x10
[  414.402442]  ? mark_held_locks+0xb0/0x110
[  414.402931]  ? __pfx_do_raw_spin_lock+0x10/0x10
[  414.403462]  ? lockdep_hardirqs_on_prepare+0x297/0x3f0
[  414.404062]  wb_workfn+0x2b3/0xcf0
[  414.404500]  ? __pfx_wb_workfn+0x10/0x10
[  414.404989]  process_scheduled_works+0x432/0x13f0
[  414.405546]  ? __pfx_process_scheduled_works+0x10/0x10
[  414.406139]  ? do_raw_spin_lock+0x101/0x2a0
[  414.406641]  ? assign_work+0x19b/0x240
[  414.407106]  ? lock_is_held_type+0x9d/0x110
[  414.407604]  worker_thread+0x6f2/0x1160
[  414.408075]  ? __kthread_parkme+0x62/0x210
[  414.408572]  ? lockdep_hardirqs_on_prepare+0x297/0x3f0
[  414.409168]  ? __kthread_parkme+0x13c/0x210
[  414.409678]  ? __pfx_worker_thread+0x10/0x10
[  414.410191]  kthread+0x33c/0x440
[  414.410602]  ? __pfx_kthread+0x10/0x10
[  414.411068]  ret_from_fork+0x4d/0x80
[  414.411526]  ? __pfx_kthread+0x10/0x10
[  414.411993]  ret_from_fork_asm+0x1b/0x30
[  414.412489]  </TASK>

When interrupt is turned on while a lock holding by spin_lock_irq it
throws a warning because of potential deadlock.

blk_mq_prep_dispatch_rq
 blk_mq_get_driver_tag
  __blk_mq_get_driver_tag
   __blk_mq_alloc_driver_tag
    blk_mq_tag_busy -> tag is already busy
    // failed to get driver tag
 blk_mq_mark_tag_wait
  spin_lock_irq(&wq->lock) -> lock A (&sbq->ws[i].wait)
  __add_wait_queue(wq, wait) -> wait queue active
  blk_mq_get_driver_tag
  __blk_mq_tag_busy
-> 1) tag must be idle, which means there can't be inflight IO
   spin_lock_irq(&tags->lock) -> lock B (hctx->tags)
   spin_unlock_irq(&tags->lock) -> unlock B, turn on interrupt accidentally
-> 2) context must be preempt by IO interrupt to trigger deadlock.

As shown above, the deadlock is not possible in theory, but the warning
still need to be fixed.

Fix it by using spin_lock_irqsave to get lockB instead of spin_lock_irq.

Fixes: 4f1731df60f9 ("blk-mq: fix potential io hang by wrong 'wake_batch'")
Signed-off-by: Li Lingfeng <[email protected]>
Reviewed-by: Ming Lei <[email protected]>
Reviewed-by: Yu Kuai <[email protected]>
Reviewed-by: Bart Van Assche <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

Merge tag 'amd-drm-fixes-6.11-2024-08-14' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.11-2024-08-14:

amdgpu:
- Fix MES ring buffer overflow
- DCN 3.5 fix
- DCN 3.2.1 fix
- DP MST fix
- Cursor fixes
- JPEG fixes
- Context ops validation
- MES 12 fixes
- VCN 5.0 fix
- HDP fix

Signed-off-by: Dave Airlie <[email protected]>
From: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

Merge tag 'perf-tools-fixes-for-v6.11-2024-08-15' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools

Pull perf tools fixes from Namhyung Kim:
 "The usual header file sync-ups and one more build fix:

   - Add README file to explain why we copy the headers

   - Sync UAPI and other header files with kernel source

   - Fix build on MIPS 32-bit"

* tag 'perf-tools-fixes-for-v6.11-2024-08-15' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools:
  perf daemon: Fix the build on 32-bit architectures
  tools/include: Sync arm64 headers with the kernel sources
  tools/include: Sync x86 headers with the kernel sources
  tools/include: Sync filesystem headers with the kernel sources
  tools/include: Sync network socket headers with the kernel sources
  tools/include: Sync uapi/asm-generic/unistd.h with the kernel sources
  tools/include: Sync uapi/sound/asound.h with the kernel sources
  tools/include: Sync uapi/linux/perf.h with the kernel sources
  tools/include: Sync uapi/linux/kvm.h with the kernel sources
  tools/include: Sync uapi/drm/i915_drm.h with the kernel sources
  perf tools: Add tools/include/uapi/README

smb: smb2pdu.h: Use static_assert() to check struct sizes

Commit 9f9bef9bc5c6 ("smb: smb2pdu.h: Avoid -Wflex-array-member-not-at-end
warnings") introduced tagged `struct create_context_hdr`. We want to
ensure that when new members need to be added to the flexible structure,
they are always included within this tagged struct.

So, we use `static_assert()` to ensure that the memory layout for
both the flexible structure and the tagged struct is the same after
any changes.

Acked-by: Namjae Jeon <[email protected]>
Signed-off-by: Gustavo A. R. Silva <[email protected]>
Signed-off-by: Steve French <[email protected]>

smb3: fix lock breakage for cached writes

Mandatory locking is enforced for cached writes, which violates
default posix semantics, and also it is enforced inconsistently.
This apparently breaks recent versions of libreoffice, but can
also be demonstrated by opening a file twice from the same
client, locking it from handle one and writing to it from
handle two (which fails, returning EACCES).

Since there was already a mount option "forcemandatorylock"
(which defaults to off), with this change only when the user
intentionally specifies "forcemandatorylock" on mount will we
break posix semantics on write to a locked range (ie we will
only fail the write in this case, if the user mounts with
"forcemandatorylock").

Fixes: 85160e03a79e ("CIFS: Implement caching mechanism for mandatory brlocks")
Cc: [email protected]
Cc: Pavel Shilovsky <[email protected]>
Reported-by: [email protected]
Reported-by: Kevin Ottens <[email protected]>
Reviewed-by: David Howells <[email protected]>
Signed-off-by: Steve French <[email protected]>

Merge tag 'md-6.11-20240815' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md into block-6.11

Pull MD fix from Song:

"This patch fixes a potential data corruption in degraded raid0 array
 with slow (WriteMostly) drives. This issue was introduced in upstream
 6.9 kernel."

* tag 'md-6.11-20240815' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md:
  md/raid1: Fix data corruption for degraded array with slow disk

md/raid1: Fix data corruption for degraded array with slow disk

read_balance() will avoid reading from slow disks as much as possible,
however, if valid data only lands in slow disks, and a new normal disk
is still in recovery, unrecovered data can be read:

raid1_read_request
 read_balance
  raid1_should_read_first
  -> return false
  choose_best_rdev
  -> normal disk is not recovered, return -1
  choose_bb_rdev
  -> missing the checking of recovery, return the normal disk
 -> read unrecovered data

Root cause is that the checking of recovery is missing in
choose_bb_rdev(). Hence add such checking to fix the problem.

Also fix similar problem in choose_slow_rdev().

Cc: [email protected]
Fixes: 9f3ced792203 ("md/raid1: factor out choose_bb_rdev() from read_balance()")
Fixes: dfa8ecd167c1 ("md/raid1: factor out choose_slow_rdev() from read_balance()")
Reported-and-tested-by: Mateusz Jończyk <[email protected]>
Closes: https://lore.kernel.org/all/[email protected]/
Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

smb/client: avoid possible NULL dereference in cifs_free_subrequest()

Clang static checker (scan-build) warning:
cifsglob.h:line 890, column 3
Access to field 'ops' results in a dereference of a null pointer.

Commit 519be989717c ("cifs: Add a tracepoint to track credits involved in
R/W requests") adds a check for 'rdata->server', and let clang throw this
warning about NULL dereference.

When 'rdata->credits.value != 0 && rdata->server == NULL' happens,
add_credits_and_wake_if() will call rdata->server->ops->add_credits().
This will cause NULL dereference problem. Add a check for 'rdata->server'
to avoid NULL dereference.

Cc: [email protected]
Fixes: 69c3c023af25 ("cifs: Implement netfslib hooks")
Reviewed-by: David Howells <[email protected]>
Signed-off-by: Su Hui <[email protected]>
Signed-off-by: Steve French <[email protected]>

Merge patch series "RISC-V: hwprobe: Misaligned scalar perf fix and rename"

Evan Green <[email protected]> says:

The CPUPERF0 hwprobe key was documented and identified in code as
a bitmask value, but its contents were an enum. This produced
incorrect behavior in conjunction with the WHICH_CPUS hwprobe flag.
The first patch in this series fixes the bitmask/enum problem by
creating a new hwprobe key that returns the same data, but is
properly described as a value instead of a bitmask. The second patch
renames the value definitions in preparation for adding vector misaligned
access info. As of this version, the old defines are kept in place to
maintain source compatibility with older userspace programs.

* b4-shazam-merge:
  RISC-V: hwprobe: Add SCALAR to misaligned perf defines
  RISC-V: hwprobe: Add MISALIGNED_PERF key

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Palmer Dabbelt <[email protected]>

riscv: Fix out-of-bounds when accessing Andes per hart vendor extension array

The out-of-bounds access is reported by UBSAN:

[    0.000000] UBSAN: array-index-out-of-bounds in ../arch/riscv/kernel/vendor_extensions.c:41:66
[    0.000000] index -1 is out of range for type 'riscv_isavendorinfo [32]'
[    0.000000] CPU: 0 UID: 0 PID: 0 Comm: swapper Not tainted 6.11.0-rc2ubuntu-defconfig #2
[    0.000000] Hardware name: riscv-virtio,qemu (DT)
[    0.000000] Call Trace:
[    0.000000] [<ffffffff94e078ba>] dump_backtrace+0x32/0x40
[    0.000000] [<ffffffff95c83c1a>] show_stack+0x38/0x44
[    0.000000] [<ffffffff95c94614>] dump_stack_lvl+0x70/0x9c
[    0.000000] [<ffffffff95c94658>] dump_stack+0x18/0x20
[    0.000000] [<ffffffff95c8bbb2>] ubsan_epilogue+0x10/0x46
[    0.000000] [<ffffffff95485a82>] __ubsan_handle_out_of_bounds+0x94/0x9c
[    0.000000] [<ffffffff94e09442>] __riscv_isa_vendor_extension_available+0x90/0x92
[    0.000000] [<ffffffff94e043b6>] riscv_cpufeature_patch_func+0xc4/0x148
[    0.000000] [<ffffffff94e035f8>] _apply_alternatives+0x42/0x50
[    0.000000] [<ffffffff95e04196>] apply_boot_alternatives+0x3c/0x100
[    0.000000] [<ffffffff95e05b52>] setup_arch+0x85a/0x8bc
[    0.000000] [<ffffffff95e00ca0>] start_kernel+0xa4/0xfb6

The dereferencing using cpu should actually not happen, so remove it.

Fixes: 23c996fc2bc1 ("riscv: Extend cpufeature.c to detect vendor extensions")
Signed-off-by: Alexandre Ghiti <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Palmer Dabbelt <[email protected]>

KEYS: trusted: dcp: fix leak of blob encryption key

Trusted keys unseal the key blob on load, but keep the sealed payload in
the blob field so that every subsequent read (export) will simply
convert this field to hex and send it to userspace.

With DCP-based trusted keys, we decrypt the blob encryption key (BEK)
in the Kernel due hardware limitations and then decrypt the blob payload.
BEK decryption is done in-place which means that the trusted key blob
field is modified and it consequently holds the BEK in plain text.
Every subsequent read of that key thus send the plain text BEK instead
of the encrypted BEK to userspace.

This issue only occurs when importing a trusted DCP-based key and
then exporting it again. This should rarely happen as the common use cases
are to either create a new trusted key and export it, or import a key
blob and then just use it without exporting it again.

Fix this by performing BEK decryption and encryption in a dedicated
buffer. Further always wipe the plain text BEK buffer to prevent leaking
the key via uninitialized memory.

Cc: [email protected] # v6.10+
Fixes: 2e8a0f40a39c ("KEYS: trusted: Introduce NXP DCP-backed trusted keys")
Signed-off-by: David Gstir <[email protected]>
Reviewed-by: Jarkko Sakkinen <[email protected]>
Signed-off-by: Jarkko Sakkinen <[email protected]>

KEYS: trusted: fix DCP blob payload length assignment

The DCP trusted key type uses the wrong helper function to store
the blob's payload length which can lead to the wrong byte order
being used in case this would ever run on big endian architectures.

Fix by using correct helper function.

Cc: [email protected] # v6.10+
Fixes: 2e8a0f40a39c ("KEYS: trusted: Introduce NXP DCP-backed trusted keys")
Suggested-by: Richard Weinberger <[email protected]>
Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Signed-off-by: David Gstir <[email protected]>
Reviewed-by: Jarkko Sakkinen <[email protected]>
Signed-off-by: Jarkko Sakkinen <[email protected]>

Merge tag 'hardening-v6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux

Pull hardening fixes from Kees Cook:

 - gcc-plugins: randstruct: Remove GCC 4.7 or newer requirement
   (Thorsten Blum)

 - kallsyms: Clean up interaction with LTO suffixes (Song Liu)

 - refcount: Report UAF for refcount_sub_and_test(0) when counter==0
   (Petr Pavlu)

 - kunit/overflow: Avoid misallocation of driver name (Ivan Orlov)

* tag 'hardening-v6.11-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  kallsyms: Match symbols exactly with CONFIG_LTO_CLANG
  kallsyms: Do not cleanup .llvm.<hash> suffix before sorting symbols
  kunit/overflow: Fix UB in overflow_allocation_test
  gcc-plugins: randstruct: Remove GCC 4.7 or newer requirement
  refcount: Report UAF for refcount_sub_and_test(0) when counter==0

btrfs: zoned: properly take lock to read/update block group's zoned variables

__btrfs_add_free_space_zoned() references and modifies bg's alloc_offset,
ro, and zone_unusable, but without taking the lock. It is mostly safe
because they monotonically increase (at least for now) and this function is
mostly called by a transaction commit, which is serialized by itself.

Still, taking the lock is a safer and correct option and I'm going to add a
change to reset zone_unusable while a block group is still alive. So, add
locking around the operations.

Fixes: 169e0da91a21 ("btrfs: zoned: track unusable bytes for zones")
CC: [email protected] # 5.15+
Reviewed-by: Johannes Thumshirn <[email protected]>
Signed-off-by: Naohiro Aota <[email protected]>
Reviewed-by: David Sterba <[email protected]>
Signed-off-by: David Sterba <[email protected]>