Add a trait that defines the DMA specific methods of devices.
The `dma::Device` trait is to be implemented by bus device
representations, where the underlying bus is capable of DMA, such as
`pci::Device` or `platform::Device`.
rust: device_id: split out index support into a separate trait
Introduce a new trait `RawDeviceIdIndex`, which extends `RawDeviceId`
to provide support for device ID types that include an index or
context field (e.g., `driver_data`). This separates the concerns of
layout compatibility and index-based data embedding, and allows
`RawDeviceId` to be implemented for types that do not contain a
`driver_data` field. Several such structures are defined in
include/linux/mod_devicetable.h.
Refactor `IdArray::new()` into a generic `build()` function, which
takes an optional offset. Based on the presence of `RawDeviceIdIndex`,
index writing is conditionally enabled. A new `new_without_index()`
constructor is also provided for use cases where no index should be
written.
This refactoring is a preparation for enabling the PHY abstractions to
use the RawDeviceId trait.
The changes to acpi.rs and driver.rs were made by Danilo.
Alice Ryhl [Fri, 11 Jul 2025 08:04:37 +0000 (08:04 +0000)]
device: rust: rename Device::as_ref() to Device::from_raw()
The prefix as_* should not be used for a constructor. Constructors
usually use the prefix from_* instead.
Some prior art in the stdlib: Box::from_raw, CString::from_raw,
Rc::from_raw, Arc::from_raw, Waker::from_raw, File::from_raw_fd.
There is also prior art in the kernel crate: cpufreq::Policy::from_raw,
fs::File::from_raw_file, Kuid::from_raw, ARef::from_raw,
SeqFile::from_raw, VmaNew::from_raw, Io::from_raw.
James Morse [Fri, 11 Jul 2025 18:27:43 +0000 (18:27 +0000)]
arm64: cacheinfo: Provide helper to compress MPIDR value into u32
Filesystems like resctrl use the cache-id exposed via sysfs to identify
groups of CPUs. The value is also used for PCIe cache steering tags. On
DT platforms cache-id is not something that is described in the
device-tree, but instead generated from the smallest MPIDR of the CPUs
associated with that cache. The cache-id exposed to user-space has
historically been 32 bits.
MPIDR values may be larger than 32 bits.
MPIDR only has 32 bits worth of affinity data, but the aff3 field lives
above 32bits. The corresponding lower bits are masked out by
MPIDR_HWID_BITMASK and contain an SMT flag and Uni-Processor flag.
Swizzzle the aff3 field into the bottom 32 bits and using that.
In case more affinity fields are added in the future, the upper RES0
area should be checked. Returning a value greater than 32 bits from
this helper will cause the caller to give up on allocating cache-ids.
Signed-off-by: James Morse <james.morse@arm.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Reviewed-by: Gavin Shan <gshan@redhat.com> Link: https://lore.kernel.org/r/20250711182743.30141-4-james.morse@arm.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
James Morse [Fri, 11 Jul 2025 18:27:42 +0000 (18:27 +0000)]
cacheinfo: Add arch hook to compress CPU h/w id into 32 bits for cache-id
Filesystems like resctrl use the cache-id exposed via sysfs to identify
groups of CPUs. The value is also used for PCIe cache steering tags. On
DT platforms cache-id is not something that is described in the
device-tree, but instead generated from the smallest CPU h/w id of the
CPUs associated with that cache.
CPU h/w ids may be larger than 32 bits.
Add a hook to allow architectures to compress the value from the devicetree
into 32 bits. Returning the same value is always safe as cache_of_set_id()
will stop if a value larger than 32 bits is seen.
For example, on arm64 the value is the MPIDR affinity register, which only
has 32 bits of affinity data, but spread accross the 64 bit field. An
arch-specific bit swizzle gives a 32 bit value.
Signed-off-by: James Morse <james.morse@arm.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Reviewed-by: Gavin Shan <gshan@redhat.com> Link: https://lore.kernel.org/r/20250711182743.30141-3-james.morse@arm.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Rob Herring [Fri, 11 Jul 2025 18:27:41 +0000 (18:27 +0000)]
cacheinfo: Set cache 'id' based on DT data
Use the minimum CPU h/w id of the CPUs associated with the cache for the
cache 'id'. This will provide a stable id value for a given system. As
we need to check all possible CPUs, we can't use the shared_cpu_map
which is just online CPUs. As there's not a cache to CPUs mapping in DT,
we have to walk all CPU nodes and then walk cache levels.
The cache_id exposed to user-space has historically been 32 bits, and
is too late to change. This value is parsed into a u32 by user-space
libraries such as libvirt:
https://github.com/libvirt/libvirt/blob/master/src/util/virresctrl.c#L1588
Give up on assigning cache-id's if a CPU h/w id greater than 32 bits
is found.
match_cache_node() does not make use of the __free() cleanup helpers
because of_find_next_cache_node(prev) does not drop a reference to prev,
and its too easy to accidentally drop the reference on cpu, which belongs
to for_each_of_cpu_node().
Cc: "Rafael J. Wysocki" <rafael@kernel.org> Signed-off-by: Rob Herring <robh@kernel.org>
[ ben: converted to use the __free cleanup idiom ] Signed-off-by: Ben Horgan <ben.horgan@arm.com>
[ morse: Add checks to give up if a value larger than 32 bits is seen. ] Signed-off-by: James Morse <james.morse@arm.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Reviewed-by: Gavin Shan <gshan@redhat.com> Link: https://lore.kernel.org/r/20250711182743.30141-2-james.morse@arm.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Sakari Ailus [Tue, 20 May 2025 10:34:37 +0000 (13:34 +0300)]
container_of: Document container_of() is not to be used in new code
There is a warning in the kerneldoc documentation of container_of() that
constness of its ptr argument is lost. While this is a valid suggestion
container_of_const() should be used instead, the vast majority of new
code still uses container_of():
Provide an unsafe functions for abstractions to convert a regular
&Device to a &Device<Bound>.
This is useful for registrations that provide certain guarantees for the
scope of their callbacks, such as IRQs or certain class device
registrations (e.g. PWM, miscdevice).
Al Viro [Wed, 2 Jul 2025 21:28:18 +0000 (22:28 +0100)]
blk-mq-debugfs: use debugfs_get_aux()
instead of manually stashing the data pointer into parent directory inode's
->i_private, just pass it to debugfs_create_file_aux() so that it can
be extracted without that insane chasing through ->d_parent.
Al Viro [Wed, 2 Jul 2025 21:25:42 +0000 (22:25 +0100)]
fix tt_command_write()
1) unbalanced debugfs_file_get(). Not needed in the first place -
file_operations are accessed only via debugfs_create_file(), so
debugfs wrappers will take care of that itself.
2) kmalloc() for a buffer used only for duration of a function is not
a problem, but for a buffer no longer than 16 bytes?
3) strstr() is for finding substrings; for finding a character there's
strchr().
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: "Rafael J. Wysocki" <rafael@kernel.org> Link: https://lore.kernel.org/r/20250702212542.GH3406663@ZenIV Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Al Viro [Wed, 2 Jul 2025 21:22:05 +0000 (22:22 +0100)]
netronome: don't bother with debugfs_real_fops()
Just turn nfp_tx_q_show() into a wrapper for helper that gets
told whether it's tx or xdp via an explicit argument and have
nfp_xdp_q_show() call the underlying helper instead.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Louis Peens <louis.peens@corigine.com> Link: https://lore.kernel.org/r/20250702212205.GF3406663@ZenIV Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Al Viro [Wed, 2 Jul 2025 21:17:39 +0000 (22:17 +0100)]
vmscan: don't bother with debugfs_real_fops()
... not when it's used only to check which file is used;
debugfs_create_file_aux_num() allows to stash a number into
debugfs entry and debugfs_get_aux_num() extracts it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Braino-spotted-by: Matthew Wilcox <willy@infradead.org> Link: https://lore.kernel.org/r/20250702211739.GE3406663@ZenIV Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Al Viro [Wed, 2 Jul 2025 21:15:08 +0000 (22:15 +0100)]
hfi1: get rid of redundant debugfs_file_{get,put}()
All files in question are created via debugfs_create_file(), so
exclusion with removals is provided by debugfs wrappers; as the matter
of fact, hfi1-private wrappers had been redundant at least since 2017...
Al Viro [Wed, 2 Jul 2025 21:14:08 +0000 (22:14 +0100)]
zynqmp: don't bother with debugfs_file_{get,put}() in proxied fops
When debugfs file has been created by debugfs_create_file_unsafe(),
we do need the file_operations methods to use debugfs_file_{get,put}()
to prevent concurrent removal; for files created by debugfs_create_file()
that is done in the wrappers that call underlying methods, so there's
no point whatsoever duplicating that in the underlying methods themselves.
Danilo Krummrich [Sat, 21 Jun 2025 19:43:33 +0000 (21:43 +0200)]
rust: pci: implement Driver::unbind()
Currently, there's really only one core callback for drivers, which is
probe().
Now, this isn't entirely true, since there is also the drop() callback of
the driver type (serving as the driver's private data), which is returned
by probe() and is dropped in remove().
On the C side remove() mainly serves two purposes:
(1) Tear down the device that is operated by the driver, e.g. call bus
specific functions, write I/O memory to reset the device, etc.
(2) Free the resources that have been allocated by a driver for a
specific device.
The drop() callback mentioned above is intended to cover (2) as the Rust
idiomatic way.
However, it is partially insufficient and inefficient to cover (1)
properly, since drop() can't be called with additional arguments, such as
the reference to the corresponding device that has the correct device
context, i.e. the Core device context.
This makes it inefficient (but not impossible) to access device
resources, e.g. to write device registers, and impossible to call device
methods, which are only accessible under the Core device context.
In order to solve this, add an additional callback for (1), which we
call unbind().
The reason for calling it unbind() is that, unlike remove(), it is *only*
meant to be used to perform teardown operations on the device (1), but
*not* to release resources (2).
Danilo Krummrich [Sat, 21 Jun 2025 19:43:32 +0000 (21:43 +0200)]
rust: platform: implement Driver::unbind()
Currently, there's really only one core callback for drivers, which is
probe().
Now, this isn't entirely true, since there is also the drop() callback of
the driver type (serving as the driver's private data), which is returned
by probe() and is dropped in remove().
On the C side remove() mainly serves two purposes:
(1) Tear down the device that is operated by the driver, e.g. call bus
specific functions, write I/O memory to reset the device, etc.
(2) Free the resources that have been allocated by a driver for a
specific device.
The drop() callback mentioned above is intended to cover (2) as the Rust
idiomatic way.
However, it is partially insufficient and inefficient to cover (1)
properly, since drop() can't be called with additional arguments, such as
the reference to the corresponding device that has the correct device
context, i.e. the Core device context.
This makes it inefficient (but not impossible) to access device
resources, e.g. to write device registers, and impossible to call device
methods, which are only accessible under the Core device context.
In order to solve this, add an additional callback for (1), which we
call unbind().
The reason for calling it unbind() is that, unlike remove(), it is *only*
meant to be used to perform teardown operations on the device (1), but
*not* to release resources (2).
Danilo Krummrich [Sat, 21 Jun 2025 19:43:27 +0000 (21:43 +0200)]
rust: device: introduce device::CoreInternal
Introduce an internal device context, which is semantically equivalent
to the Core device context, but reserved for bus abstractions.
This allows implementing methods for the Device type, which are limited
to be used within the core context of bus abstractions, i.e. restrict
the availability for drivers.
Miguel Ojeda [Tue, 1 Jul 2025 17:46:56 +0000 (19:46 +0200)]
rust: acpi: remove unneeded cast to clean future Clippy warning
A future Clippy warning, `clippy::as_underscore`, is getting enabled in
parallel in the rust-next tree:
error: using `as _` conversion
--> rust/kernel/acpi.rs:25:9
|
25 | self.0.driver_data as _
| ^^^^^^^^^^^^^^^^^^^^^^-
| |
| help: consider giving the type explicitly: `usize`
The type is already `ulong`, which nowadays is always `usize`, so the
cast is unneeded. Thus remove it, which in turn will avoid the warning
in the future.
Other abstractions of device tables do not use a cast here either.
Matti Vaittinen [Mon, 9 Jun 2025 07:26:47 +0000 (10:26 +0300)]
device property: Use tidy for_each_named_* macros
Implementing if-conditions inside for_each_x() macros requires some
thinking to avoid side effects in the calling code. Resulting code
may look somewhat awkward, and there are couple of different ways it is
usually done.
Standardizing this to one way can help making it more obvious for a code
reader and writer. The newly added for_each_if() is a way to achieve this.
Use for_each_if() to make these macros look like many others which
should in the long run help reading the code.
Shankari Anand [Thu, 26 Jun 2025 10:45:20 +0000 (16:15 +0530)]
rust: miscdevice: clarify invariant for `MiscDeviceRegistration`
Reword and expand the invariant documentation for `MiscDeviceRegistration`
to clarify what it means for the inner device to be "registered".
It expands to explain:
- `inner` points to a `miscdevice` registered via `misc_register`.
- This registration stays valid for the entire lifetime of the object.
- Deregistration is guaranteed on `Drop`, via `misc_deregister`.
Danilo Krummrich [Thu, 26 Jun 2025 20:00:41 +0000 (22:00 +0200)]
rust: devres: get rid of Devres' inner Arc
So far Devres uses an inner memory allocation and reference count, i.e.
an inner Arc, in order to ensure that the devres callback can't run into
a use-after-free in case where the Devres object is dropped while the
devres callback runs concurrently.
Instead, use a completion in order to avoid a potential UAF: In
Devres::drop(), if we detect that we can't remove the devres action
anymore, we wait for the completion that is completed from the devres
callback. If, in turn, we were able to successfully remove the devres
action, we can just go ahead.
This, again, allows us to get rid of the internal Arc, and instead let
Devres consume an `impl PinInit<T, E>` in order to return an
`impl PinInit<Devres<T>, E>`, which enables us to get away with less
memory allocations.
Additionally, having the resulting explicit synchronization in
Devres::drop() prevents potential subtle undesired side effects of the
devres callback dropping the final Arc reference asynchronously within
the devres callback.
Danilo Krummrich [Thu, 26 Jun 2025 20:00:40 +0000 (22:00 +0200)]
rust: devres: replace Devres::new_foreign_owned()
Replace Devres::new_foreign_owned() with devres::register().
The current implementation of Devres::new_foreign_owned() creates a full
Devres container instance, including the internal Revocable and
completion.
However, none of that is necessary for the intended use of giving full
ownership of an object to devres and getting it dropped once the given
device is unbound.
Hence, implement devres::register(), which is limited to consume the
given data, wrap it in a KBox and drop the KBox once the given device is
unbound, without any other synchronization.
Moves the implementation for `pin-init` from an associated function
to the trait function of the `Wrapper` trait and extends the
implementation to support pin-initializers with error types.
Adds a use for the `Wrapper` trait in `revocable.rs`, to use the new
`pin-init` function. This is currently the only usage in the kernel.
Reviewed-by: Gerald Wisböck <gerald.wisboeck@feather.ink> Reviewed-by: Alice Ryhl <aliceryhl@google.com> Reviewed-by: Benno Lossin <lossin@kernel.org> Acked-by: Miguel Ojeda <ojeda@kernel.org> Signed-off-by: Christian Schrefl <chrisi.schrefl@gmail.com> Link: https://lore.kernel.org/r/20250610-b4-rust_miscdevice_registrationdata-v6-1-b03f5dfce998@gmail.com Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Igor Korotin [Fri, 20 Jun 2025 15:43:34 +0000 (16:43 +0100)]
rust: platform: Add ACPI match table support to `Driver` trait
Extend the `platform::Driver` trait to support ACPI device matching by
adding the `ACPI_ID_TABLE` constant.
This allows Rust platform drivers to define ACPI match tables alongside
their existing OF match tables. These changes mirror the existing OF
support and allow Rust platform drivers to match devices based on ACPI
identifiers.
Igor Korotin [Fri, 20 Jun 2025 15:39:13 +0000 (16:39 +0100)]
rust: driver: Add ACPI id table support to Adapter trait
Extend the `Adapter` trait to support ACPI device identification.
This mirrors the existing Open Firmware (OF) support (`of_id_table`) and
enables Rust drivers to match and retrieve ACPI-specific device data
when `CONFIG_ACPI` is enabled.
To avoid breaking compilation, a stub implementation of `acpi_id_table()`
is added to the Platform adapter; the full implementation will be provided
in a subsequent patch.
Igor Korotin [Fri, 20 Jun 2025 15:36:56 +0000 (16:36 +0100)]
rust: driver: Consolidate `Adapter::of_id_info` methods using `#[cfg]`
Refactor the `of_id_info` methods in the `Adapter` trait to reduce
duplication. Previously, the method had two versions selected
via `#[cfg(...)]` and `#[cfg(not(...))]`. This change merges them into a
single method by using `#[cfg]` blocks within the method body.
Suggested-by: Benno Lossin <lossin@kernel.org> Signed-off-by: Igor Korotin <igor.korotin.linux@gmail.com> Link: https://lore.kernel.org/r/20250620153656.294468-1-igor.korotin.linux@gmail.com
[ Fix clippy warning if #[cfg(not(CONFIG_OF))]; fix checkpatch.pl line
length warnings. - Danilo ] Signed-off-by: Danilo Krummrich <dakr@kernel.org>
samples: rust: platform: Add property child and reference args examples
Add some example usage of the device property methods for reading
DT/ACPI/swnode child nodes and reference args.
Signed-off-by: Remo Senekowitsch <remo@buenzli.dev> Link: https://lore.kernel.org/r/20250616154511.1862909-4-remo@buenzli.dev
[ Convert 'child@{0,1}' to 'child-{0,1}'; skip child nodes without
'compatible' property in of_unittest_platform_populate() as proposed
by Rob Herring. - Danilo] Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Allow Rust code to read reference args from device properties. The
wrapper type `FwNodeReferenceArgs` allows callers to access the buffer
of read args safely.
Allow Rust drivers to access children of a fwnode either by name or by
iterating over all of them.
In C, there is the function `fwnode_get_next_child_node` for iteration
and the macro `fwnode_for_each_child_node` that helps with handling the
pointers. Instead of a macro, a native iterator is used in Rust such
that regular for-loops can be used.
Linus Torvalds [Sun, 22 Jun 2025 17:50:36 +0000 (10:50 -0700)]
Merge tag 'i2c-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
- subsystem: convert drivers to use recent callbacks of struct
i2c_algorithm A typical after-rc1 cleanup, which I couldn't send in
time for rc2
- tegra: fix YAML conversion of device tree bindings
- k1: re-add a check which got lost during upstreaming
* tag 'i2c-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: k1: check for transfer error
i2c: use inclusive callbacks in struct i2c_algorithm
dt-bindings: i2c: nvidia,tegra20-i2c: Specify the required properties
Linus Torvalds [Sun, 22 Jun 2025 17:30:44 +0000 (10:30 -0700)]
Merge tag 'x86_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Borislav Petkov:
- Make sure the array tracking which kernel text positions need to be
alternatives-patched doesn't get mishandled by out-of-order
modifications, leading to it overflowing and causing page faults when
patching
- Avoid an infinite loop when early code does a ranged TLB invalidation
before the broadcast TLB invalidation count of how many pages it can
flush, has been read from CPUID
- Fix a CONFIG_MODULES typo
- Disable broadcast TLB invalidation when PTI is enabled to avoid an
overflow of the bitmap tracking dynamic ASIDs which need to be
flushed when the kernel switches between the user and kernel address
space
- Handle the case of a CPU going offline and thus reporting zeroes when
reading top-level events in the resctrl code
* tag 'x86_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/alternatives: Fix int3 handling failure from broken text_poke array
x86/mm: Fix early boot use of INVPLGB
x86/its: Fix an ifdef typo in its_alloc()
x86/mm: Disable INVLPGB when PTI is enabled
x86,fs/resctrl: Remove inappropriate references to cacheinfo in the resctrl subsystem
Linus Torvalds [Sun, 22 Jun 2025 17:17:51 +0000 (10:17 -0700)]
Merge tag 'irq_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq fixes from Borislav Petkov:
- Fix missing prototypes warnings
- Properly initialize work context when allocating it
- Remove a method tracking when managed interrupts are suspended during
hotplug, in favor of the code using a IRQ disable depth tracking now,
and have interrupts get properly enabled again on restore
- Make sure multiple CPUs getting hotplugged don't cause wrong tracking
of the managed IRQ disable depth
* tag 'irq_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip/ath79-misc: Fix missing prototypes warnings
genirq/irq_sim: Initialize work context pointers properly
genirq/cpuhotplug: Restore affinity even for suspended IRQ
genirq/cpuhotplug: Rebalance managed interrupts across multi-CPU hotplug
Linus Torvalds [Sun, 22 Jun 2025 17:11:45 +0000 (10:11 -0700)]
Merge tag 'perf_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Borislav Petkov:
- Avoid a crash on a heterogeneous machine where not all cores support
the same hw events features
- Avoid a deadlock when throttling events
- Document the perf event states more
- Make sure a number of perf paths switching off or rescheduling events
call perf_cgroup_event_disable()
- Make sure perf does task sampling before its userspace mapping is
torn down, and not after
* tag 'perf_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel: Fix crash in icl_update_topdown_event()
perf: Fix the throttle error of some clock events
perf: Add comment to enum perf_event_state
perf/core: Fix WARN in perf_cgroup_switch()
perf: Fix dangling cgroup pointer in cpuctx
perf: Fix cgroup state vs ERROR
perf: Fix sample vs do_exit()
Linus Torvalds [Sun, 22 Jun 2025 17:09:23 +0000 (10:09 -0700)]
Merge tag 'locking_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking fixes from Borislav Petkov:
- Make sure the switch to the global hash is requested always under a
lock so that two threads requesting that simultaneously cannot get to
inconsistent state
- Reject negative NUMA nodes earlier in the futex NUMA interface
handling code
- Selftests fixes
* tag 'locking_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Verify under the lock if hash can be replaced
futex: Handle invalid node numbers supplied by user
selftests/futex: Set the home_node in futex_numa_mpol
selftests/futex: getopt() requires int as return value.
* tag 'edac_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
EDAC/igen6: Fix NULL pointer dereference
EDAC/amd64: Correct number of UMCs for family 19h models 70h-7fh
Linus Torvalds [Sun, 22 Jun 2025 16:46:11 +0000 (09:46 -0700)]
Merge tag 'v6.16-rc2-smb3-client-fixes-v2' of git://git.samba.org/sfrench/cifs-2.6
Pull smb client fixes from Steve French:
- Multichannel channel allocation fix for Kerberos mounts
- Two reconnect fixes
- Fix netfs_writepages crash with smbdirect/RDMA
- Directory caching fix
- Three minor cleanup fixes
- Log error when close cached dirs fails
* tag 'v6.16-rc2-smb3-client-fixes-v2' of git://git.samba.org/sfrench/cifs-2.6:
smb: minor fix to use SMB2_NTLMV2_SESSKEY_SIZE for auth_key size
smb: minor fix to use sizeof to initialize flags_string buffer
smb: Use loff_t for directory position in cached_dirents
smb: Log an error when close_all_cached_dirs fails
cifs: Fix prepare_write to negotiate wsize if needed
smb: client: fix max_sge overflow in smb_extract_folioq_to_rdma()
smb: client: fix first command failure during re-negotiation
cifs: Remove duplicate fattr->cf_dtype assignment from wsl_to_fattr() function
smb: fix secondary channel creation issue with kerberos by populating hostname when adding channels
Alex Elder [Mon, 16 Jun 2025 12:51:36 +0000 (07:51 -0500)]
i2c: k1: check for transfer error
If spacemit_i2c_xfer_msg() times out waiting for a message transfer to
complete, or if the hardware reports an error, it returns a negative
error code (-ETIMEDOUT, -EAGAIN, -ENXIO. or -EIO).
The sole caller of spacemit_i2c_xfer_msg() is spacemit_i2c_xfer(),
which is the i2c_algorithm->xfer callback function. It currently
does not save the value returned by spacemit_i2c_xfer_msg().
The result is that transfer errors go unreported, and a caller
has no indication anything is wrong.
When this code was out for review, the return value *was* checked
in early versions. But for some reason, that assignment got dropped
between versions 5 and 6 of the series, perhaps related to reworking
the code to merge spacemit_i2c_xfer_core() into spacemit_i2c_xfer().
Simply assigning the value returned to "ret" fixes the problem.
Fixes: 5ea558473fa31 ("i2c: spacemit: add support for SpacemiT K1 SoC") Signed-off-by: Alex Elder <elder@riscstar.com> Cc: <stable@vger.kernel.org> # v6.15+ Reviewed-by: Troy Mitchell <troymitchell988@gmail.com> Link: https://lore.kernel.org/r/20250616125137.1555453-1-elder@riscstar.com Signed-off-by: Andi Shyti <andi@smida.it> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Linus Torvalds [Sat, 21 Jun 2025 16:20:15 +0000 (09:20 -0700)]
Merge tag 'nfsd-6.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux
Pull nfsd fixes from Chuck Lever:
- Two fixes for commits in the nfsd-6.16 merge
- One fix for the recently-added NFSD netlink facility
- One fix for a remote SunRPC crasher
* tag 'nfsd-6.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux:
sunrpc: handle SVC_GARBAGE during svc auth processing as auth error
nfsd: use threads array as-is in netlink interface
SUNRPC: Cleanup/fix initial rq_pages allocation
NFSD: Avoid corruption of a referring call list
Bharath SM [Thu, 19 Jun 2025 15:35:34 +0000 (21:05 +0530)]
smb: minor fix to use SMB2_NTLMV2_SESSKEY_SIZE for auth_key size
Replaced hardcoded value 16 with SMB2_NTLMV2_SESSKEY_SIZE
in the auth_key definition and memcpy call.
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Signed-off-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Bharath SM [Thu, 19 Jun 2025 15:35:33 +0000 (21:05 +0530)]
smb: minor fix to use sizeof to initialize flags_string buffer
Replaced hardcoded length with sizeof(flags_string).
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Signed-off-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Bharath SM [Thu, 19 Jun 2025 15:35:32 +0000 (21:05 +0530)]
smb: Use loff_t for directory position in cached_dirents
Change the pos field in struct cached_dirents from int to loff_t
to support large directory offsets. This avoids overflow and
matches kernel conventions for directory positions.
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Signed-off-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Paul Aurich [Wed, 20 Nov 2024 16:01:54 +0000 (08:01 -0800)]
smb: Log an error when close_all_cached_dirs fails
Under low-memory conditions, close_all_cached_dirs() can't move the
dentries to a separate list to dput() them once the locks are dropped.
This will result in a "Dentry still in use" error, so add an error
message that makes it clear this is what happened:
[ 495.281119] CIFS: VFS: \\otters.example.com\share Out of memory while dropping dentries
[ 495.281595] ------------[ cut here ]------------
[ 495.281887] BUG: Dentry ffff888115531138{i=78,n=/} still in use (2) [unmount of cifs cifs]
[ 495.282391] WARNING: CPU: 1 PID: 2329 at fs/dcache.c:1536 umount_check+0xc8/0xf0
Also, bail out of looping through all tcons as soon as a single
allocation fails, since we're already in trouble, and kmalloc() attempts
for subseqeuent tcons are likely to fail just like the first one did.
Signed-off-by: Paul Aurich <paul@darkrain42.org> Acked-by: Bharath SM <bharathsm@microsoft.com> Suggested-by: Ruben Devos <rdevos@oxya.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
David Howells [Wed, 18 Jun 2025 15:39:47 +0000 (16:39 +0100)]
cifs: Fix prepare_write to negotiate wsize if needed
Fix cifs_prepare_write() to negotiate the wsize if it is unset.
Reviewed-by: Shyam Prasad N <nspmangalore@gmail.com> Reviewed-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: David Howells <dhowells@redhat.com>
cc: Paulo Alcantara <pc@manguebit.org>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-cifs@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
cc: Tom Talpey <tom@talpey.com>
cc: linux-cifs@vger.kernel.org Reviewed-by: David Howells <dhowells@redhat.com> Reviewed-by: Tom Talpey <tom@talpey.com> Fixes: c45ebd636c32 ("cifs: Provide the capability to extract from ITER_FOLIOQ to RDMA SGEs") Signed-off-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Steve French <stfrench@microsoft.com>
zhangjian [Thu, 19 Jun 2025 01:18:29 +0000 (09:18 +0800)]
smb: client: fix first command failure during re-negotiation
after fabc4ed200f9, server_unresponsive add a condition to check whether client
need to reconnect depending on server->lstrp. When client failed to reconnect
for some time and abort connection, server->lstrp is updated for the last time.
In the following scene, server->lstrp is too old. This cause next command
failure in re-negotiation rather than waiting for re-negotiation done.
1. mount -t cifs -o username=Everyone,echo_internal=10 //$server_ip/export /mnt
2. ssh $server_ip "echo b > /proc/sysrq-trigger &"
3. ls /mnt
4. sleep 21s
5. ssh $server_ip "service firewalld stop"
6. ls # return EHOSTDOWN
If the interval between 5 and 6 is too small, 6 may trigger sending negotiation
request. Before backgrounding cifsd thread try to receive negotiation response
from server in cifs_readv_from_socket, server_unresponsive may trigger
cifs_reconnect which cause 6 to be failed:
ls thread
----------------
smb2_negotiate
server->tcpStatus = CifsInNegotiate
compound_send_recv
wait_for_compound_request
ls thread
----------------
cifs_sync_mid_result return EAGAIN
smb2_negotiate return EHOSTDOWN
Though server->lstrp means last server response time, it is updated in
cifs_abort_connection and cifs_get_tcp_session. We can also update server->lstrp
before switching into CifsInNegotiate state to avoid failure in 6.
Fixes: 7ccc1465465d ("smb: client: fix hang in wait_for_response() for negproto") Acked-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Acked-by: Meetakshi Setiya <msetiya@microsoft.com> Signed-off-by: zhangjian <zhangjian496@huawei.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Linus Torvalds [Sat, 21 Jun 2025 15:27:12 +0000 (08:27 -0700)]
Merge tag 'acpi-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI fix from Rafael Wysocki:
"Fix a crash in ACPICA while attempting to evaluate a control method
that expects more arguments than are being passed to it, which was
exposed by a defective firmware update from a prominent OEM on
multiple systems (Rafael Wysocki)"
* tag 'acpi-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPICA: Refuse to evaluate a method if arguments are missing
Linus Torvalds [Sat, 21 Jun 2025 15:21:10 +0000 (08:21 -0700)]
Merge tag 'pci-v6.16-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci
Pull PCI fixes from Bjorn Helgaas:
- Set up runtime PM even for devices that lack a PM Capability as we
did before 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when
initializing"), which broke resume in some VFIO scenarios (Mario
Limonciello)
- Ignore pciehp Presence Detect Changed events caused by DPC, even if
they occur after a Data Link Layer State Changed event, to fix a VFIO
GPU passthrough regression in v6.13 (Lukas Wunner)
* tag 'pci-v6.16-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
PCI: pciehp: Ignore belated Presence Detect Changed caused by DPC
PCI/PM: Set up runtime PM even for devices without PCI PM
Linus Torvalds [Sat, 21 Jun 2025 14:59:45 +0000 (07:59 -0700)]
Merge tag 'perf-tools-fixes-for-v6.16-1-2025-06-20' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools
Pull perf tools fixes from Arnaldo Carvalho de Melo:
- Fix some file descriptor leaks that stand out with recent changes to
'perf list'
- Fix prctl include to fix building 'perf bench futex' hash with musl
libc
- Restrict 'perf test' uniquifying entry to machines with 'uncore_imc'
PMUs
- Document new output fields (op, cache, mem, dtlb, snoop) used with
'perf mem'
- Synchronize kernel header copies
* tag 'perf-tools-fixes-for-v6.16-1-2025-06-20' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools:
tools headers x86 cpufeatures: Sync with the kernel sources
perf bench futex: Fix prctl include in musl libc
perf test: Directory file descriptor leak
perf evsel: Missed close() when probing hybrid core PMUs
tools headers: Synchronize linux/bits.h with the kernel sources
tools arch amd ibs: Sync ibs.h with the kernel sources
tools arch x86: Sync the msr-index.h copy with the kernel sources
tools headers: Syncronize linux/build_bug.h with the kernel sources
tools headers: Update the copy of x86's mem{cpy,set}_64.S used in 'perf bench'
tools headers UAPI: Sync linux/kvm.h with the kernel sources
tools headers UAPI: Sync the drm/drm.h with the kernel sources
perf beauty: Update copy of linux/socket.h with the kernel sources
tools headers UAPI: Sync kvm header with the kernel sources
tools headers x86 svm: Sync svm headers with the kernel sources
tools headers UAPI: Sync KVM's vmx.h header with the kernel sources
tools kvm headers arm64: Update KVM header from the kernel sources
tools headers UAPI: Sync linux/prctl.h with the kernel sources to pick FUTEX knob
perf mem: Document new output fields (op, cache, mem, dtlb, snoop)
tools headers: Update the fs headers with the kernel sources
perf test: Restrict uniquifying test to machines with 'uncore_imc'
Linus Torvalds [Sat, 21 Jun 2025 05:36:48 +0000 (22:36 -0700)]
Merge tag 'mtd/fixes-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux
Pull mtd fixes from Miquel Raynal:
"The main fix that really needs to get in is the revert of the patch
adding the new mtd_master class, because it entirely fails the
partitioning if a specific Kconfig option is set. We need to think how
to handle that differently, so let's revert it as we need to get back
to the pen and paper situation again.
Otherwise the definition of some Winbond SPI NAND chips are receiving
some fixes (geometry and maximum frequency, mostly).
And finally a small memory leak gets also fixed"
* tag 'mtd/fixes-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux:
mtd: spinand: fix memory leak of ECC engine conf
mtd: spinand: winbond: Prevent unsupported frequencies on dual/quad I/O variants
mtd: spinand: winbond: Increase maximum frequency on an octal operation
mtd: spinand: winbond: Fix W35N number of planes/LUN
Revert "mtd: core: always create master device"
Currently the call_rcu() API does not check whether a callback
pointer is NULL. If NULL is passed, rcu_core() will try to invoke
it, resulting in NULL pointer dereference and a kernel crash.
To prevent this and improve debuggability, this patch adds a check
for NULL and emits a kernel stack trace to help identify a faulty
caller.
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Reviewed-by: Joel Fernandes <joelagnelf@nvidia.com> Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
Binbin Wu [Tue, 10 Jun 2025 02:14:21 +0000 (10:14 +0800)]
KVM: TDX: Exit to userspace for GetTdVmCallInfo
Exit to userspace for TDG.VP.VMCALL<GetTdVmCallInfo> via KVM_EXIT_TDX,
to allow userspace to provide information about the support of
TDVMCALLs when r12 is 1 for the TDVMCALLs beyond the GHCI base API.
GHCI spec defines the GHCI base TDVMCALLs: <GetTdVmCallInfo>, <MapGPA>,
<ReportFatalError>, <Instruction.CPUID>, <#VE.RequestMMIO>,
<Instruction.HLT>, <Instruction.IO>, <Instruction.RDMSR> and
<Instruction.WRMSR>. They must be supported by VMM to support TDX guests.
For GetTdVmCallInfo
- When leaf (r12) to enumerate TDVMCALL functionality is set to 0,
successful execution indicates all GHCI base TDVMCALLs listed above are
supported.
Update the KVM TDX document with the set of the GHCI base APIs.
- When leaf (r12) to enumerate TDVMCALL functionality is set to 1, it
indicates the TDX guest is querying the supported TDVMCALLs beyond
the GHCI base TDVMCALLs.
Exit to userspace to let userspace set the TDVMCALL sub-function bit(s)
accordingly to the leaf outputs. KVM could set the TDVMCALL bit(s)
supported by itself when the TDVMCALLs don't need support from userspace
after returning from userspace and before entering guest. Currently, no
such TDVMCALLs implemented, KVM just sets the values returned from
userspace.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
[Adjust userspace API. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Binbin Wu [Tue, 10 Jun 2025 02:14:20 +0000 (10:14 +0800)]
KVM: TDX: Handle TDG.VP.VMCALL<GetQuote>
Handle TDVMCALL for GetQuote to generate a TD-Quote.
GetQuote is a doorbell-like interface used by TDX guests to request VMM
to generate a TD-Quote signed by a service hosting TD-Quoting Enclave
operating on the host. A TDX guest passes a TD Report (TDREPORT_STRUCT) in
a shared-memory area as parameter. Host VMM can access it and queue the
operation for a service hosting TD-Quoting enclave. When completed, the
Quote is returned via the same shared-memory area.
KVM only checks the GPA from the TDX guest has the shared-bit set and drops
the shared-bit before exiting to userspace to avoid bleeding the shared-bit
into KVM's exit ABI. KVM forwards the request to userspace VMM (e.g. QEMU)
and userspace VMM queues the operation asynchronously. KVM sets the return
code according to the 'ret' field set by userspace to notify the TDX guest
whether the request has been queued successfully or not. When the request
has been queued successfully, the TDX guest can poll the status field in
the shared-memory area to check whether the Quote generation is completed
or not. When completed, the generated Quote is returned via the same
buffer.
Add KVM_EXIT_TDX as a new exit reason to userspace. Userspace is
required to handle the KVM exit reason as the initial support for TDX,
by reentering KVM to ensure that the TDVMCALL is complete. While at it,
add a note that KVM_EXIT_HYPERCALL also requires reentry with KVM_RUN.
Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com> Tested-by: Mikko Ylinen <mikko.ylinen@linux.intel.com> Acked-by: Kai Huang <kai.huang@intel.com>
[Adjust userspace API. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Binbin Wu [Tue, 10 Jun 2025 02:14:19 +0000 (10:14 +0800)]
KVM: TDX: Add new TDVMCALL status code for unsupported subfuncs
Add the new TDVMCALL status code TDVMCALL_STATUS_SUBFUNC_UNSUPPORTED and
return it for unimplemented TDVMCALL subfunctions.
Returning TDVMCALL_STATUS_INVALID_OPERAND when a subfunction is not
implemented is vague because TDX guests can't tell the error is due to
the subfunction is not supported or an invalid input of the subfunction.
New GHCI spec adds TDVMCALL_STATUS_SUBFUNC_UNSUPPORTED to avoid the
ambiguity. Use it instead of TDVMCALL_STATUS_INVALID_OPERAND.
Before the change, for common guest implementations, when a TDX guest
receives TDVMCALL_STATUS_INVALID_OPERAND, it has two cases:
1. Some operand is invalid. It could change the operand to another value
retry.
2. The subfunction is not supported.
For case 1, an invalid operand usually means the guest implementation bug.
Since the TDX guest can't tell which case is, the best practice for
handling TDVMCALL_STATUS_INVALID_OPERAND is stopping calling such leaf,
treating the failure as fatal if the TDVMCALL is essential or ignoring
it if the TDVMCALL is optional.
With this change, TDVMCALL_STATUS_SUBFUNC_UNSUPPORTED could be sent to
old TDX guest that do not know about it, but it is expected that the
guest will make the same action as TDVMCALL_STATUS_INVALID_OPERAND.
Currently, no known TDX guest checks TDVMCALL_STATUS_INVALID_OPERAND
specifically; for example Linux just checks for success.
Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
[Return it for untrapped KVM_HC_MAP_GPA_RANGE. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Linus Torvalds [Fri, 20 Jun 2025 16:59:20 +0000 (09:59 -0700)]
Merge tag 'sound-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"A collection of small fixes. All changes are device-specific at this
time:
- Fixes for Cirrus codecs with SoundWire, including firmware name
updates
- Fix for i.MX8 SoC DSP
- Usual HD-audio, USB-audio, and ASoC AMD quirks
- Fixes for legendary SoundBlaster AWE32 ISA device (a real one, we
still got a bug report after 25 years)
- Minor build fixes"
* tag 'sound-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (24 commits)
ALSA: hda/realtek: Enable headset Mic on Positivo P15X
ASoC: Intel: sof-function-topology-lib: Print out the unsupported dmic count
ASoC: doc: cs35l56: Add CS35L63 to the list of supported devices
ASoC: SOF: imx8: add core shutdown operation for imx8/imx8x
ALSA: hda/realtek: Add quirk for Asus GA605K
ALSA: hda/realtek: enable headset mic on Latitude 5420 Rugged
ASoC: amd: yc: update quirk data for HP Victus
ASoC: apple: mca: Drop default ARCH_APPLE in Kconfig
ALSA: usb-audio: Rename ALSA kcontrol PCM and PCM1 for the KTMicro sound card
ASoC: amd: yc: Add quirk for MSI Bravo 17 D7VF internal mic
ASoC: doc: cs35l56: Update to add new SoundWire firmware filename suffix
ASoC: cs35l56: Use SoundWire address as alternate firmware suffix on L56 B0
ASoC: cs35l56: Use SoundWire address as firmware name suffix for new silicon
ASoC: sdw_utils: Fix potential NULL pointer deref in is_sdca_endpoint_present()
ALSA: sb: Force to disable DMAs once when DMA mode is changed
ALSA: sb: Don't allow changing the DMA mode during operations
ALSA: hda/realtek: Add quirk for Asus GU605C
ALSA: hda/realtek: Fix built-in mic on ASUS VivoBook X513EA
ALSA: hda/realtek - Add mute LED support for HP Victus 16-s1xxx and HP Victus 15-fa1xxx
ALSA: ctxfi: Replace deprecated strcpy() with strscpy()
...
Linus Torvalds [Fri, 20 Jun 2025 16:54:24 +0000 (09:54 -0700)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"There's nothing major (even the vmalloc one is just suppressing a
potential warning) but all worth having, nonetheless.
- Suppress KASAN false positive in stack unwinding code
- Drop redundant reset of the GCS state on exec()
- Don't try to descend into a !present PMD when creating a huge
vmap() entry at the PUD level
- Fix a small typo in the arm64 booting Documentation"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64/ptrace: Fix stack-out-of-bounds read in regs_get_kernel_stack_nth()
arm64/gcs: Don't call gcs_free() during flush_gcs()
arm64: Restrict pagetable teardown to avoid false warning
docs: arm64: Fix ICC_SRE_EL2 register typo in booting.rst
projects / thirdparty / kernel / stable.git / log
