linux-2.6-microblaze.git
5 weeks agofirewire: core: fix undefined reference error in ARM EABI
Takashi Sakamoto [Sun, 28 Sep 2025 01:19:10 +0000 (10:19 +0900)]
firewire: core: fix undefined reference error in ARM EABI

For ARM EABI, GCC generates a reference to __aeabi_uldivmod when compiling
a division of 64-bit integer with 32-bit integer. This function is not
available in Linux kernel. In such cases, helper macros are defined in
include/linux/math64.h.

This commit replaces the division with div_u64().

Fixes: 8ec6a8ec23b9 ("firewire: core: suppress overflow warning when computing jiffies from isochronous cycle")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202509270428.FZaO2PPq-lkp@intel.com/
Link: https://lore.kernel.org/r/20250928011910.581475-1-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agoRevert "firewire: core: disable bus management work temporarily during updating topology"
Takashi Sakamoto [Wed, 24 Sep 2025 13:18:23 +0000 (22:18 +0900)]
Revert "firewire: core: disable bus management work temporarily during updating topology"

This reverts commit abe7159125702c734e851bc0c52b51cd446298a5.

The bus manager work item acquires the spin lock of fw_card again, thus
no need to serialize it against fw_core_handle_bus_reset().

Link: https://lore.kernel.org/r/20250924131823.262136-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agoRevert "firewire: core: shrink critical section of fw_card spinlock in bm_work"
Takashi Sakamoto [Wed, 24 Sep 2025 13:18:22 +0000 (22:18 +0900)]
Revert "firewire: core: shrink critical section of fw_card spinlock in bm_work"

This reverts commit 582310376d6e9a8d261b682178713cdc4b251af6.

The bus manager work has the race condition against fw_destroy_nodes()
called by fw_core_remove_card(). The acquition of spin lock of fw_card
is left as is again.

Link: https://lore.kernel.org/r/20250924131823.262136-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core: suppress overflow warning when computing jiffies from isochronous...
Takashi Sakamoto [Wed, 24 Sep 2025 13:11:40 +0000 (22:11 +0900)]
firewire: core: suppress overflow warning when computing jiffies from isochronous cycle

The multiplication by USEC_PER_SEC (=1000000L) may trigger an overflow
warning with 32 bit storage. In the case of the subsystem the input value
ranges between 800 and 16000, thus the result always fits within 32 bit
storage.

This commit suppresses the warning by using widening conversion to 64 bit
storage before multiplication, then using narrowing conversion to 32 bit
storage.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202509170136.b5ZHaNAV-lkp@intel.com/
Fixes: 379b870c28c6 ("firewire: core: use helper macros instead of direct access to HZ")
Link: https://lore.kernel.org/r/20250924131140.261686-1-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core: minor code refactoring to delete useless local variable
Takashi Sakamoto [Thu, 18 Sep 2025 23:54:48 +0000 (08:54 +0900)]
firewire: core: minor code refactoring to delete useless local variable

The do_reset local variable has less merit. Let's remove it.

Link: https://lore.kernel.org/r/20250918235448.129705-7-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core; eliminate pick_me goto label
Takashi Sakamoto [Thu, 18 Sep 2025 23:54:47 +0000 (08:54 +0900)]
firewire: core; eliminate pick_me goto label

This commit uses condition statements instead of pick_me goto label.

Link: https://lore.kernel.org/r/20250918235448.129705-6-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core: code refactoring to split contention procedure for bus manager
Takashi Sakamoto [Thu, 18 Sep 2025 23:54:46 +0000 (08:54 +0900)]
firewire: core: code refactoring to split contention procedure for bus manager

The precedure to contend for bus manager has much code. It is better to
split it into a helper function.

This commit refactors in the point.

Link: https://lore.kernel.org/r/20250918235448.129705-5-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core: code refactoring for the case of generation mismatch
Takashi Sakamoto [Thu, 18 Sep 2025 23:54:45 +0000 (08:54 +0900)]
firewire: core: code refactoring for the case of generation mismatch

Current implementation stores the bus generation at which the bus manager
contending procedure finishes. The condition for the procedure is the
mismatch of the stored generation against current bus generation.

This commit refactors the code for the contending procedure. Two existing
branches are put into a new branch to detect the generation mismatch, thus
the most of change is indentation.

Link: https://lore.kernel.org/r/20250918235448.129705-4-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core: use switch statement to evaluate transaction result to CSR_BUS_MANAGER_ID
Takashi Sakamoto [Thu, 18 Sep 2025 23:54:44 +0000 (08:54 +0900)]
firewire: core: use switch statement to evaluate transaction result to CSR_BUS_MANAGER_ID

The result of the lock transaction to swap bus manager on isochronous
resource manager looks like an ad-hoc style. It is hard to read.

This commit uses switch statement to evaluate the result.

Link: https://lore.kernel.org/r/20250918235448.129705-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core: remove useless generation check
Takashi Sakamoto [Thu, 18 Sep 2025 23:54:43 +0000 (08:54 +0900)]
firewire: core: remove useless generation check

Two functions, fw_core_handle_bus_reset() and bm_work(), are serialized
by a commit 3d91fd440cc7 ("firewire: core: disable bus management work
temporarily during updating topology"). Therefore the generation member
of fw_card is immutable in bm_work().

This commit removes useless generation check in bm_work().

Link: https://lore.kernel.org/r/20250918235448.129705-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
6 weeks agofirewire: core: use struct_size and flex_array_size in ioctl_add_descriptor
Thorsten Blum [Tue, 16 Sep 2025 12:21:45 +0000 (14:21 +0200)]
firewire: core: use struct_size and flex_array_size in ioctl_add_descriptor

Use struct_size() to determine the memory needed for a new 'struct
descriptor_resource' and flex_array_size() to calculate the number of
bytes to copy from userspace. This removes the hardcoded size (4 bytes)
for the 'u32 data[]' entries.

No functional changes intended.

Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Link: https://lore.kernel.org/r/20250916122143.2459993-3-thorsten.blum@linux.dev
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: shrink critical section of fw_card spinlock in bm_work
Takashi Sakamoto [Wed, 17 Sep 2025 00:03:47 +0000 (09:03 +0900)]
firewire: core: shrink critical section of fw_card spinlock in bm_work

Now fw_core_handle_bus_reset() and bm_work() are serialized. Some members
of fw_card are free to access in bm_work()

This commit shrinks critical section of fw_card spinlock in bm_work()

Link: https://lore.kernel.org/r/20250917000347.52369-4-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: disable bus management work temporarily during updating topology
Takashi Sakamoto [Wed, 17 Sep 2025 00:03:46 +0000 (09:03 +0900)]
firewire: core: disable bus management work temporarily during updating topology

When processing selfID sequence, bus topology tree is (re)built, and some
members of fw_card are determined. Once determined, the members are valid
during the bus generation. The above operations are in the critical
section of fw_card spin lock.

Before building the bus topology, a work item is scheduled for bus manager
work. The bm_work() function is invoked by the work item. The function
tries to acquire the spin lock, then can be stalled until the bus topology
building finishes.

The bus manager should work once the members of fw_card are determined.
This commit suppresses the above situation by disabling the work item
during processing selfID sequence.

Link: https://lore.kernel.org/r/20250917000347.52369-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: schedule bm_work item outside of spin lock
Takashi Sakamoto [Wed, 17 Sep 2025 00:03:45 +0000 (09:03 +0900)]
firewire: core: schedule bm_work item outside of spin lock

Before (re)building topology tree, fw_core_handle_bus_reset() schedules
a work item under acquiring fw_card spin lock. The work item invokes
bm_work() which acquires the spin lock at first, then can be stalled to
wait until the building tree finishes. This is inconvenient.

This commit moves the timing to schedule the work item after releasing
the spin lock.

Link: https://lore.kernel.org/r/20250917000347.52369-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: annotate fw_destroy_nodes with must-hold-lock
Takashi Sakamoto [Mon, 15 Sep 2025 23:47:47 +0000 (08:47 +0900)]
firewire: core: annotate fw_destroy_nodes with must-hold-lock

The function, fw_destroy_nodes(), is used widely within firewire-core
module. It has a prerequisite condition that struct fw_card.lock must
be hold in advance.

This commit adds annotation for it.

Link: https://lore.kernel.org/r/20250915234747.915922-7-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use spin lock specific to timer for split transaction
Takashi Sakamoto [Mon, 15 Sep 2025 23:47:46 +0000 (08:47 +0900)]
firewire: core: use spin lock specific to timer for split transaction

At present the parameters to compute timeout time for split transaction is
protected by card-wide spin lock, while it is not necessarily convenient
in a point to narrower critical section.

This commit adds and uses another spin lock specific for the purpose.

Link: https://lore.kernel.org/r/20250915234747.915922-6-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use spin lock specific to transaction
Takashi Sakamoto [Mon, 15 Sep 2025 23:47:45 +0000 (08:47 +0900)]
firewire: core: use spin lock specific to transaction

The list of instance for asynchronous transaction to wait for response
subaction is maintained as a member of fw_card structure. The card-wide
spinlock is used at present for any operation over the list, however it
is not necessarily suited for the purpose.

This commit adds and uses the spin lock specific to maintain the list.

Link: https://lore.kernel.org/r/20250915234747.915922-5-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use spin lock specific to topology map
Takashi Sakamoto [Mon, 15 Sep 2025 23:47:44 +0000 (08:47 +0900)]
firewire: core: use spin lock specific to topology map

At present, the operation for read transaction to topology map register is
not protected by any kind of lock primitives. This causes a potential
problem to result in the mixed content of topology map.

This commit adds and uses spin lock specific to topology map.

Link: https://lore.kernel.org/r/20250915234747.915922-4-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: maintain phy packet receivers locally in cdev layer
Takashi Sakamoto [Mon, 15 Sep 2025 23:47:43 +0000 (08:47 +0900)]
firewire: core: maintain phy packet receivers locally in cdev layer

The list of receivers for phy packet is used only by cdev layer, while it
is maintained as a member of fw_card structure.

This commit maintains the list locally in cdev layer.

Link: https://lore.kernel.org/r/20250915234747.915922-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use scoped_guard() to manage critical section to update topology
Takashi Sakamoto [Mon, 15 Sep 2025 23:47:42 +0000 (08:47 +0900)]
firewire: core: use scoped_guard() to manage critical section to update topology

At present, guard() macro is used for the critical section to update
topology. It is inconvenient to add the other critical sections into
the function.

This commit uses scoped_guard() macro instead.

Link: https://lore.kernel.org/r/20250915234747.915922-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use helper macro to compare against current jiffies
Takashi Sakamoto [Mon, 15 Sep 2025 02:42:32 +0000 (11:42 +0900)]
firewire: core: use helper macro to compare against current jiffies

The pattern of calling either time_before64() or time_after64() with
get_jiffies_64() can be replaced with the corresponding helper macros.

Link: https://lore.kernel.org/r/20250915024232.851955-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use helper macros instead of direct access to HZ
Takashi Sakamoto [Mon, 15 Sep 2025 02:42:31 +0000 (11:42 +0900)]
firewire: core: use helper macros instead of direct access to HZ

There are some macros available to convert usecs, msecs, and secs into
jiffies count.

Link: https://lore.kernel.org/r/20250915024232.851955-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use macro expression for not-registered state of BUS_MANAGER_ID
Takashi Sakamoto [Sat, 13 Sep 2025 10:57:37 +0000 (19:57 +0900)]
firewire: core: use macro expression for not-registered state of BUS_MANAGER_ID

The value of BUS_MANAGER_ID register has 0x3f when no node_id is
registered. Current implementation uses hard-coded numeric literal but
in the case the macro expression is preferable since it is easy to
distinguish the state from node ID mask.

This commit applies the idea.

Link: https://lore.kernel.org/r/20250913105737.778038-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: use macro expression for gap count mismatch
Takashi Sakamoto [Sat, 13 Sep 2025 10:57:36 +0000 (19:57 +0900)]
firewire: core: use macro expression for gap count mismatch

The gap_count field is assigned to zero when mismatch is detected. In such
case, the macro expression is preferable since it is easy to understand
the situation.

This commit applies the idea.

Link: https://lore.kernel.org/r/20250913105737.778038-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
7 weeks agofirewire: core: remove useless lockdep_assert_held()
Takashi Sakamoto [Thu, 11 Sep 2025 22:13:11 +0000 (07:13 +0900)]
firewire: core: remove useless lockdep_assert_held()

The bm_work work item should be scheduled after holding fw_card reference
counting. At a commit 25feb1a96e21 ("firewire: core: use cleanup function
in bm_work"), I misinterpreted it as fw_card spinlock and inserted
lockdep_assert_hold() wrongly.

This commit removes the useless line.

Fixes: 25feb1a96e21 ("firewire: core: use cleanup function in bm_work")
Link: https://lore.kernel.org/r/20250911221312.678076-1-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: code refactoring whether root node is cycle master capable
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:08 +0000 (10:21 +0900)]
firewire: core: code refactoring whether root node is cycle master capable

The check of cycle master capability in root node is currently just in a
condition branch. In this case, the required variable should be within the
branch.

This commit is just for the purpose.

Link: https://lore.kernel.org/r/20250908012108.514698-12-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: code refactoring to investigate root node for bus manager
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:07 +0000 (10:21 +0900)]
firewire: core: code refactoring to investigate root node for bus manager

In the middle of bm_work function, both the value of gap_count and the
state of root node are investigated. Current implementation is not a good
shape since the investigation is aligned to be flat.

This commit refactors the investigation with two large branches.

Link: https://lore.kernel.org/r/20250908012108.514698-11-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: code refactoring to detect both IEEE 1394:1995 IRM and Canon MV5i
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:06 +0000 (10:21 +0900)]
firewire: core: code refactoring to detect both IEEE 1394:1995 IRM and Canon MV5i

The detection of IEEE 1394:1995 and Canon MV5i is just required within
some of the condition branches. In this case, these check can be
capsulated within these branches.

This commit refactors the checks.

Link: https://lore.kernel.org/r/20250908012108.514698-10-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: refer fw_card member to initiate bus reset under acquiring lock
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:05 +0000 (10:21 +0900)]
firewire: core: refer fw_card member to initiate bus reset under acquiring lock

The gap_count member of fw_card structure is referred when initiate bus
reset. This reference is done out of acquiring lock. This is not good.

This commit takes the reference within the acquiring lock, with
additional code refactoring.

Link: https://lore.kernel.org/r/20250908012108.514698-9-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: code refactoring to evaluate transaction result to CSR_BUS_MANAGER_ID
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:04 +0000 (10:21 +0900)]
firewire: core: code refactoring to evaluate transaction result to CSR_BUS_MANAGER_ID

The call of bm_work should be done after acquiring spin lock of fw_card.
For asynchronous transaction, the lock should be released temporarily
due to event waiting.

A commit 27310d561622 ("firewire: core: use guard macro to maintain
properties of fw_card") applied scoped_guard() to the bm_work function,
however it looks hard to follow to the control flow.

This commit refactors the spin lock acquisition after the transaction.

Link: https://lore.kernel.org/r/20250908012108.514698-8-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: ohci: localize transaction data and rcode per condition branch
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:03 +0000 (10:21 +0900)]
firewire: ohci: localize transaction data and rcode per condition branch

The function local variable, transaction_data, in bm_work function is
conditionally used. In the case, the branch-level variable is sometimes
useful.

This commit uses this idea.

Link: https://lore.kernel.org/r/20250908012108.514698-7-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: use cleanup function in bm_work
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:02 +0000 (10:21 +0900)]
firewire: core: use cleanup function in bm_work

In "bm_work" function, the references to fw_card and fw_node are
released at last. This is achieved by using goto statements. For this
case, the kernel cleanup framework is available.

This commit uses the framework to remove these statements.

Link: https://lore.kernel.org/r/20250908012108.514698-6-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: add helper functions to access to fw_device data in fw_node structure
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:01 +0000 (10:21 +0900)]
firewire: core: add helper functions to access to fw_device data in fw_node structure

The data mbmer in fw_node structure is an opaque pointer, while nowadays
it is just used to refer to fw_device associated with the fw_node.

This commit redefines the opaque pointer to a pointer to fw_device
structure, and adds some helper functions to set/get it.

Link: https://lore.kernel.org/r/20250908012108.514698-5-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: ohci: use return value from fw_node_get()
Takashi Sakamoto [Mon, 8 Sep 2025 01:21:00 +0000 (10:21 +0900)]
firewire: ohci: use return value from fw_node_get()

The programming pattern, referring after increasing reference count, is
supported by fw_node_get().

This commit simplify the programming pattern.

Link: https://lore.kernel.org/r/20250908012108.514698-4-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: core: utilize cleanup function to release workqueue in error path
Takashi Sakamoto [Mon, 8 Sep 2025 01:20:59 +0000 (10:20 +0900)]
firewire: core: utilize cleanup function to release workqueue in error path

The helper macro, retain_and_null_ptr(), introduced by a commit
092d00ead733 ("cleanup: Provide retain_and_null_ptr()") in v6.16 kernel,
is useful in the error path to release the part of structure member.

This commit uses the relatively new function.

Link: https://lore.kernel.org/r/20250908012108.514698-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
8 weeks agofirewire: ohci: use kcalloc() variant for array allocation
Takashi Sakamoto [Mon, 8 Sep 2025 01:20:58 +0000 (10:20 +0900)]
firewire: ohci: use kcalloc() variant for array allocation

When allocating the list of isochronous context structure, a kzalloc()
variant of managed device API is used. In this case, a kcalloc() variant
is available.

This commit replaces these lines with devm_kcalloc().

Link: https://lore.kernel.org/r/20250908012108.514698-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agofirewire: ohci: remove module-local workqueue
Takashi Sakamoto [Sat, 23 Aug 2025 03:09:54 +0000 (12:09 +0900)]
firewire: ohci: remove module-local workqueue

Now module-local workqueue has been replaced by a threaded IRQ handler.

This commit removes the workqueue and the associated work item
accordingly.

Link: https://lore.kernel.org/r/20250823030954.268412-4-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agofirewire: ohci: use threaded IRQ handler to handle SelfIDComplete event
Takashi Sakamoto [Sat, 23 Aug 2025 03:09:53 +0000 (12:09 +0900)]
firewire: ohci: use threaded IRQ handler to handle SelfIDComplete event

The first step  maintaining the bus topology is to handle SelfIDComplete
event. This event occurs after initiating bus reset when 1394 OHCI link
layer is enabled, or when the bus topology changes (e.g. when a device is
added). Because enumeration of the selfID sequence can take some time, it
should be processed in a bottom half.

Currently, this is done in a module-local workqueue with the
WQ_MEM_RECLAIM flag, to allow invocation during memory reclaim paths. A
threaded IRQ handler is a preferable alternative, as it eliminates the
need to manage workqueue attributes manually.

Although SelfIDComplete events are not so frequent in normal usage,
handling them correctly is critical for proper bus topology management.
This commit switches SelfIDComplete handling to a threaded IRQ handler.

Link: https://lore.kernel.org/r/20250823030954.268412-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agofirewire: ohci: move self_id_complete tracepoint after validating register
Takashi Sakamoto [Sat, 23 Aug 2025 03:09:52 +0000 (12:09 +0900)]
firewire: ohci: move self_id_complete tracepoint after validating register

The value of OHCI1394_SelfIDCount register includes an error-indicating
bit. It is safer to place the tracepoint probe after validating the
register value.

Link: https://lore.kernel.org/r/20250823030954.268412-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agofirewire: ohci: remove obsolete module-level debug parameter
Takashi Sakamoto [Thu, 21 Aug 2025 00:30:17 +0000 (09:30 +0900)]
firewire: ohci: remove obsolete module-level debug parameter

The module-level debug parameter was added in v2.6.26 by a commit
ad3c0fe8b8d16 ("firewire: debug interrupt events"). Its functionality
has long been superseded by tracepoints.

This commit removes the module parameter, bye.

Link: https://lore.kernel.org/r/20250821003017.186752-5-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agofirewire: ohci: remove obsolete debug logging for AT/AR results
Takashi Sakamoto [Thu, 21 Aug 2025 00:30:16 +0000 (09:30 +0900)]
firewire: ohci: remove obsolete debug logging for AT/AR results

Between v6.11 and v6.12, a set of tracepoints was added to record
asynchronous communication events:

- firewire:async_phy_inbound
- firewire:async_phy_outbound_initiate
- firewire:async_phy_outbound_complete
- firewire:async_response_inbound
- firewire:async_response_outbound_initiate
- firewire:async_response_outbound_complete
- firewire:async_request_inbound
- firewire:async_request_outbound_initiate
- firewire:async_request_outbound_complete

These tracepoints cover the functionality of the existing debug logging.
This commit removes the logging.

Link: https://lore.kernel.org/r/20250821003017.186752-4-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agofirewire: ohci: remove obsolete debug logging for selfID sequence
Takashi Sakamoto [Thu, 21 Aug 2025 00:30:15 +0000 (09:30 +0900)]
firewire: ohci: remove obsolete debug logging for selfID sequence

A commit 677ceae19073 ("firewire: core: add tracepoints event for
self_id_sequence") added the "firewire:self_id_sequence" event in v6.11.
A commit 526e21a2aa6f ("firewire: ohci: add tracepoints event for data
of Self-ID DMA") added the "firewire_ohci:self_id_complete" event in
v6.12.

These tracepoints replace the equivalent debug logging. This commit
removes the logging.

Link: https://lore.kernel.org/r/20250821003017.186752-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agofirewire: ohci: remove obsolete debug logging for IRQ events
Takashi Sakamoto [Thu, 21 Aug 2025 00:30:14 +0000 (09:30 +0900)]
firewire: ohci: remove obsolete debug logging for IRQ events

A commit 0d8914165dd1 ("firewire: ohci: add tracepoints event for hardIRQ
event") added "firewire_ohci:irqs" event in v6.11, which can provide
equivalent information to the existing debug logging.

This commit removes the logging.

Link: https://lore.kernel.org/r/20250821003017.186752-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
2 months agoLinux 6.17-rc2
Linus Torvalds [Sun, 17 Aug 2025 22:22:10 +0000 (15:22 -0700)]
Linux 6.17-rc2

2 months agoMerge tag 'x86_urgent_for_v6.17_rc2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 17 Aug 2025 13:53:15 +0000 (06:53 -0700)]
Merge tag 'x86_urgent_for_v6.17_rc2' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Borislav Petkov:

 - Remove a transitional asm/cpuid.h header which was added only as a
   fallback during cpuid helpers reorg

 - Initialize reserved fields in the SVSM page validation calls
   structure to zero in order to allow for future structure extensions

 - Have the sev-guest driver's buffers used in encryption operations be
   in linear mapping space as the encryption operation can be offloaded
   to an accelerator

 - Have a read-only MSR write when in an AMD SNP guest trap to the
   hypervisor as it is usually done. This makes the guest user
   experience better by simply raising a #GP instead of terminating said
   guest

 - Do not output AVX512 elapsed time for kernel threads because the data
   is wrong and fix a NULL pointer dereferencing in the process

 - Adjust the SRSO mitigation selection to the new attack vectors

* tag 'x86_urgent_for_v6.17_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/cpuid: Remove transitional <asm/cpuid.h> header
  x86/sev: Ensure SVSM reserved fields in a page validation entry are initialized to zero
  virt: sev-guest: Satisfy linear mapping requirement in get_derived_key()
  x86/sev: Improve handling of writes to intercepted TSC MSRs
  x86/fpu: Fix NULL dereference in avx512_status()
  x86/bugs: Select best SRSO mitigation

2 months agoMerge tag 'locking_urgent_for_v6.17_rc2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 17 Aug 2025 12:57:47 +0000 (05:57 -0700)]
Merge tag 'locking_urgent_for_v6.17_rc2' of git://git./linux/kernel/git/tip/tip

Pull locking fixes from Borislav Petkov:

 - Make sure sanity checks down in the mutex lock path happen on the
   correct type of task so that they don't trigger falsely

 - Use the write unsafe user access pairs when writing a futex value to
   prevent an error on PowerPC which does user read and write accesses
   differently

* tag 'locking_urgent_for_v6.17_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking: Fix __clear_task_blocked_on() warning from __ww_mutex_wound() path
  futex: Use user_write_access_begin/_end() in futex_put_value()

2 months agoMerge tag 'rust-fixes-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda...
Linus Torvalds [Sat, 16 Aug 2025 22:49:24 +0000 (15:49 -0700)]
Merge tag 'rust-fixes-6.17' of git://git./linux/kernel/git/ojeda/linux

Pull rust fixes from Miguel Ojeda:

 - Workaround 'rustdoc' target modifiers bug in Rust >= 1.88.0. It will
   be fixed in Rust 1.90.0 (expected 2025-09-18).

 - Clean 'rustdoc' output before running it to avoid confusing the tool
   when files from previous versions remain.

* tag 'rust-fixes-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux:
  rust: kbuild: clean output before running `rustdoc`
  rust: workaround `rustdoc` target modifiers bug

2 months agoMerge tag 'ata-ata-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/libata...
Linus Torvalds [Sat, 16 Aug 2025 13:59:13 +0000 (06:59 -0700)]
Merge tag 'ata-ata-6.17-rc2' of git://git./linux/kernel/git/libata/linux

Pull ata fixes from Damien Le Moal:

 - Fix a regression affecting old IDE/PATA device scan and introduced by
   the recent link power management cleanups & fixes. The regression
   prevented devices from being properly detected (me)

 - Fix command duration limits (CDL) feature control: attempting to
   enable the feature while NCQ commands are being executed resulted in
   a silent failure to enable CDL when needed (Igor)

* tag 'ata-ata-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux:
  ata: libata-scsi: Fix CDL control
  ata: libata-eh: Fix link state check for IDE/PATA ports

2 months agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Sat, 16 Aug 2025 13:20:49 +0000 (06:20 -0700)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "One core change removing the 'w' access flag of attributes that don't
  have a set routine (and therefore can't be written to) which should
  have no practical impact. The big scsi_debug update is caused by
  reformatting lots of arrays and the rest of the bug fixes in drivers
  are trivial"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: ufs: core: Remove error print for devm_add_action_or_reset()
  scsi: ufs: mediatek: Fix out-of-bounds access in MCQ IRQ mapping
  scsi: lpfc: Remove redundant assignment to avoid memory leak
  scsi: lpfc: Fix wrong function reference in a comment
  scsi: ufs: core: Fix interrupt handling for MCQ Mode
  scsi: scsi_debug: Make read-only arrays static const
  scsi: core: sysfs: Correct sysfs attributes access rights

2 months agoMerge tag 'drm-fixes-2025-08-16' of https://gitlab.freedesktop.org/drm/kernel
Linus Torvalds [Fri, 15 Aug 2025 21:52:36 +0000 (14:52 -0700)]
Merge tag 'drm-fixes-2025-08-16' of https://gitlab.freedesktop.org/drm/kernel

Pull drm fixes from Dave Airlie:
 "Relatively quiet week, usual amdgpu/i915/xe fixes along with a set of
  fixes for fbdev format info, which fix some regressions seen in with
  rc1.

  bridge:
   - fix OF-node leak
   - fix documentation

  fbdev-emulation:
   - pass correct format info to drm_helper_mode_fill_fb_struct()

  panfrost:
   - print correct RSS size

  amdgpu:
   - PSP fix
   - VRAM reservation fix
   - CSA fix
   - Process kill fix

  i915:
   - Fix the implementation of wa_18038517565 [fbc]
   - Do not trigger Frame Change events from frontbuffer flush [psr]

  xe:
   - Some more xe_migrate_access_memory fixes (Auld)
   - Defer buffer object shrinker write-backs and GPU waits (Thomas)
   - HWMON fix for clamping limits (Karthik)
   - SRIOV-PF: Set VF LMEM BAR size (Michal)"

* tag 'drm-fixes-2025-08-16' of https://gitlab.freedesktop.org/drm/kernel:
  drm/xe/pf: Set VF LMEM BAR size
  drm/amdgpu: fix task hang from failed job submission during process kill
  drm/amdgpu: fix incorrect vm flags to map bo
  drm/amdgpu: fix vram reservation issue
  drm/amdgpu: Add PSP fw version check for fw reserve GFX command
  drm/xe/hwmon: Add SW clamp for power limits writes
  drm/xe: Defer buffer object shrinker write-backs and GPU waits
  drm/xe/migrate: prevent potential UAF
  drm/xe/migrate: don't overflow max copy size
  drm/xe/migrate: prevent infinite recursion
  drm/i915/psr: Do not trigger Frame Change events from frontbuffer flush
  drm/i915/fbc: fix the implementation of wa_18038517565
  drm/panfrost: Print RSS for tiler heap BO's in debugfs GEMS file
  drm/radeon: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()
  drm/nouveau: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()
  drm/omap: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()
  drm/bridge: document HDMI CEC callbacks
  drm/bridge: Describe the newly introduced drm_connector parameter for drm_bridge_detect
  drm/bridge: fix OF node leak

2 months agoMerge tag 'xfs-fixes-6.17-rc2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Linus Torvalds [Fri, 15 Aug 2025 19:50:12 +0000 (12:50 -0700)]
Merge tag 'xfs-fixes-6.17-rc2' of git://git./fs/xfs/xfs-linux

Pull xfs fixes from Carlos Maiolino:

 - Fix an assert trigger introduced during the merge window

 - Prevent atomic writes to be used with DAX

 - Prevent users from using the max_atomic_write mount option without
   reflink, as atomic writes > 1block are not supported without reflink

 - Fix a null-pointer-deref in a tracepoint

* tag 'xfs-fixes-6.17-rc2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: split xfs_zone_record_blocks
  xfs: fix scrub trace with null pointer in quotacheck
  xfs: reject max_atomic_write mount option for no reflink
  xfs: disallow atomic writes on DAX
  fs/dax: Reject IOCB_ATOMIC in dax_iomap_rw()
  xfs: remove XFS_IBULK_SAME_AG
  xfs: fully decouple XFS_IBULK* flags from XFS_IWALK* flags
  xfs: fix frozen file system assert in xfs_trans_alloc

2 months agoMerge tag 'block-6.17-20250815' of git://git.kernel.dk/linux
Linus Torvalds [Fri, 15 Aug 2025 15:20:36 +0000 (08:20 -0700)]
Merge tag 'block-6.17-20250815' of git://git.kernel.dk/linux

Pull block fixes from Jens Axboe:

 - Fix for unprivileged daemons in ublk

 - Speedup ublk release by removing unnecessary quiesce

 - Fix for blk-wbt, where a regression caused it to not be possible to
   enable at runtime

 - blk-wbt cleanups

 - Kill the page pool from drbd

 - Remove redundant __GFP_NOWARN uses in a few spots

 - Fix for a kobject double initialization issues

* tag 'block-6.17-20250815' of git://git.kernel.dk/linux:
  block: restore default wbt enablement
  Docs: admin-guide: Correct spelling mistake
  blk-wbt: doc: Update the doc of the wbt_lat_usec interface
  blk-wbt: Eliminate ambiguity in the comments of struct rq_wb
  blk-wbt: Optimize wbt_done() for non-throttled writes
  block: fix kobject double initialization in add_disk
  blk-cgroup: remove redundant __GFP_NOWARN
  block, bfq: remove redundant __GFP_NOWARN
  ublk: check for unprivileged daemon on each I/O fetch
  ublk: don't quiesce in ublk_ch_release
  drbd: Remove the open-coded page pool

2 months agox86/cpuid: Remove transitional <asm/cpuid.h> header
Ahmed S. Darwish [Fri, 15 Aug 2025 07:01:54 +0000 (09:01 +0200)]
x86/cpuid: Remove transitional <asm/cpuid.h> header

All CPUID call sites were updated at commit:

    968e30006807 ("x86/cpuid: Set <asm/cpuid/api.h> as the main CPUID header")

to include <asm/cpuid/api.h> instead of <asm/cpuid.h>.

The <asm/cpuid.h> header was still retained as a wrapper, just in case
some new code in -next started using it.  Now that everything is merged
to Linus' tree, remove the header.

Signed-off-by: Ahmed S. Darwish <darwi@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/20250815070227.19981-2-darwi@linutronix.de
2 months agox86/sev: Ensure SVSM reserved fields in a page validation entry are initialized to...
Tom Lendacky [Wed, 13 Aug 2025 15:26:59 +0000 (10:26 -0500)]
x86/sev: Ensure SVSM reserved fields in a page validation entry are initialized to zero

In order to support future versions of the SVSM_CORE_PVALIDATE call, all
reserved fields within a PVALIDATE entry must be set to zero as an SVSM should
be ensuring all reserved fields are zero in order to support future usage of
reserved areas based on the protocol version.

Fixes: fcd042e86422 ("x86/sev: Perform PVALIDATE using the SVSM when not at VMPL0")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Joerg Roedel <joerg.roedel@amd.com>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/7cde412f8b057ea13a646fb166b1ca023f6a5031.1755098819.git.thomas.lendacky@amd.com
2 months agovirt: sev-guest: Satisfy linear mapping requirement in get_derived_key()
Tom Lendacky [Wed, 16 Jul 2025 20:41:35 +0000 (15:41 -0500)]
virt: sev-guest: Satisfy linear mapping requirement in get_derived_key()

Commit

  7ffeb2fc2670 ("x86/sev: Document requirement for linear mapping of guest request buffers")

added a check that requires the guest request buffers to be in the linear
mapping. The get_derived_key() function was passing a buffer that was
allocated on the stack, resulting in the call to snp_send_guest_request()
returning an error.

Update the get_derived_key() function to use an allocated buffer instead
of a stack buffer.

Fixes: 7ffeb2fc2670 ("x86/sev: Document requirement for linear mapping of guest request buffers")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/9b764ca9fc79199a091aac684c4926e2080ca7a8.1752698495.git.thomas.lendacky@amd.com
2 months agoMerge tag 'io_uring-6.17-20250815' of git://git.kernel.dk/linux
Linus Torvalds [Fri, 15 Aug 2025 15:02:34 +0000 (08:02 -0700)]
Merge tag 'io_uring-6.17-20250815' of git://git.kernel.dk/linux

Pull io_uring fixes from Jens Axboe:

 - Tweak for the fairly recent changes of minimizing io-wq worker
   creations when it's pointless to create them.

 - Fix for an issue with ring provided buffers, which could cause issues
   with reuse or corrupt application data.

* tag 'io_uring-6.17-20250815' of git://git.kernel.dk/linux:
  io_uring/io-wq: add check free worker before create new worker
  io_uring/net: commit partial buffers on retry

2 months agoMerge tag 'sound-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 15 Aug 2025 14:02:57 +0000 (07:02 -0700)]
Merge tag 'sound-6.17-rc2' of git://git./linux/kernel/git/tiwai/sound

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

   - Potential OOB access fixes in USB-audio driver

   - ASoC kconfig menu fix for improving the generic drivers

   - HD-audio quirks and a fix revert

   - Codec and platform-specific small fixes for ASoC"

* tag 'sound-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda/tas2781: Normalize the volume kcontrol name
  ALSA: usb-audio: Validate UAC3 cluster segment descriptors
  ALSA: usb-audio: Validate UAC3 power domain descriptors, too
  Revert "ALSA: hda: Add ASRock X670E Taichi to denylist"
  ALSA: azt3328: Put __maybe_unused for inline functions for gameport
  ASoC: tas2781: Normalize the volume kcontrol name
  ASoC: stm: stm32_i2s: Fix calc_clk_div() error handling in determine_rate()
  ASoC: codecs: Call strscpy() with correct size argument
  ALSA: hda/realtek: Fix headset mic on HONOR BRB-X
  ALSA: hda/realtek: Add Framework Laptop 13 (AMD Ryzen AI 300) to quirks
  ASoC: tas2781: Fix spelling mistake "dismatch" -> "mismatch"
  ASoC: rt1320: fix random cycle mute issue
  ASoC: rt721: fix FU33 Boost Volume control not working
  ASoC: generic: tidyup standardized ASoC menu for generic
  ASoC: codec: sma1307: replace spelling mistake with new error message
  ASoC: codecs: tx-macro: correct tx_macro_component_drv name
  ASoC: fsl_sai: replace regmap_write with regmap_update_bits

2 months agoMerge tag 'gpio-fixes-for-v6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 15 Aug 2025 13:54:37 +0000 (06:54 -0700)]
Merge tag 'gpio-fixes-for-v6.17-rc2' of git://git./linux/kernel/git/brgl/linux

Pull gpio fix from Bartosz Golaszewski:

 - fix the way optional interrupts are retrieved from firmware in
   gpio-mlxbf3

* tag 'gpio-fixes-for-v6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpio: mlxbf3: use platform_get_irq_optional()
  Revert "gpio: mlxbf3: only get IRQ for device instance 0"

2 months agoMerge tag 'pmdomain-v6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh...
Linus Torvalds [Fri, 15 Aug 2025 13:30:53 +0000 (06:30 -0700)]
Merge tag 'pmdomain-v6.17-rc1' of git://git./linux/kernel/git/ulfh/linux-pm

Pull pmdomain fix from Ulf Hansson:

 - tegra: Ensure pmc power-domains are in a known state

* tag 'pmdomain-v6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm:
  soc/tegra: pmc: Ensure power-domains are in a known state

2 months agoMerge tag '6.17-rc1-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Linus Torvalds [Fri, 15 Aug 2025 12:58:19 +0000 (05:58 -0700)]
Merge tag '6.17-rc1-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6

Pull smb client fixes from Steve French:

 - Fix unlink race and rename races

 - SMB3.1.1 compression fix

 - Avoid unneeded strlen calls in cifs_get_spnego_key

 - Fix slab out of bounds in parse_server_interfaces()

 - Fix mid leak and server buffer leak

 - smbdirect send error path fix

 - update internal version #

 - Fix unneeded response time update in negotiate protocol

* tag '6.17-rc1-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
  smb: client: remove redundant lstrp update in negotiate protocol
  cifs: update internal version number
  smb: client: don't wait for info->send_pending == 0 on error
  smb: client: fix mid_q_entry memleak leak with per-mid locking
  smb3: fix for slab out of bounds on mount to ksmbd
  cifs: avoid extra calls to strlen() in cifs_get_spnego_key()
  cifs: Fix collect_sample() to handle any iterator type
  smb: client: fix race with concurrent opens in rename(2)
  smb: client: fix race with concurrent opens in unlink(2)

2 months agoMerge tag 'firewire-fixes-6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 15 Aug 2025 02:15:22 +0000 (19:15 -0700)]
Merge tag 'firewire-fixes-6.17-rc1' of git://git./linux/kernel/git/ieee1394/linux1394

Pull firewire fixes from Takashi Sakamoto:
 "This fixes a potential call to schedule() within an RCU read-side
  critical section. The solution applies reference counting to ensure
  that handlers which may call schedule() are invoked safely outside of
  the critical section"

* tag 'firewire-fixes-6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:
  firewire: core: reallocate buffer for FCP address handlers when more than 4 are registered
  firewire: core: call FCP address handlers outside RCU read-side critical section
  firewire: core: call handler for exclusive regions outside RCU read-side critical section
  firewire: core: use reference counting to invoke address handlers safely

2 months agoMerge tag 'drm-xe-fixes-2025-08-14' of https://gitlab.freedesktop.org/drm/xe/kernel...
Dave Airlie [Thu, 14 Aug 2025 23:50:17 +0000 (09:50 +1000)]
Merge tag 'drm-xe-fixes-2025-08-14' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes

- Some more xe_migrate_access_memory fixes (Auld)
- Defer buffer object shrinker write-backs and GPU waits (Thomas)
- HWMON fix for clamping limits (Karthik)
- SRIOV-PF: Set VF LMEM BAR size (Michal)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/aJ4MIZQurSo0uNxn@intel.com
2 months agoMerge tag 'drm-intel-fixes-2025-08-13' of https://gitlab.freedesktop.org/drm/i915...
Dave Airlie [Thu, 14 Aug 2025 23:05:04 +0000 (09:05 +1000)]
Merge tag 'drm-intel-fixes-2025-08-13' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-fixes

- Fix the implementation of wa_18038517565 [fbc] (Vinod Govindapillai)
- Do not trigger Frame Change events from frontbuffer flush [psr] (Jouni Högander)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Tvrtko Ursulin <tursulin@igalia.com>
Link: https://lore.kernel.org/r/aJ0HAh06VHWVdv63@linux
2 months agoMerge tag 'acpi-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Thu, 14 Aug 2025 15:20:58 +0000 (08:20 -0700)]
Merge tag 'acpi-6.17-rc2' of git://git./linux/kernel/git/rafael/linux-pm

Pull ACPI fixes from Rafael Wysocki:
 "These restore corner case behavior of the EC driver related to the
  handling of defective ACPI tables and fix a recent regression in the
  ACPI processor driver:

   - Prevent the ACPI EC driver from ignoring ECDT information in the
     cases when the ID string in the ECDT is invalid, but not empty, to
     fix thouchpad detection on ThinkBook 14 G7 IML (Armin Wolf)

   - Rearrange checks in acpi_processor_ppc_init() to restore the
     handling of frequency QoS requests related to _PPC limits
     inadvertently broken by a recent update (Rafael Wysocki)"

* tag 'acpi-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: EC: Relax sanity check of the ECDT ID string
  ACPI: processor: perflib: Move problematic pr->performance check

2 months agoMerge tag 'pm-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Linus Torvalds [Thu, 14 Aug 2025 14:55:31 +0000 (07:55 -0700)]
Merge tag 'pm-6.17-rc2' of git://git./linux/kernel/git/rafael/linux-pm

Pull power management fixes from Rafael Wysocki:
 "These remove an artificial limitation from the intel_idle driver,
  update the menu cpuidle governor to restore its previous behavior in a
  corner case and add one more supported platform configuration to the
  intel_pstate driver:

   - Allow intel_idle to use _CST information from ACPI tables for idle
     states enumeration on any family of processors (Len Brown)

   - Restore corner case behavior of the menu cpuidle governor, related
     to the handling of systems where idle states selected by the
     governor are rejected by the cpuidle driver, inadvertently changed
     during the 6.15 development cycle (Rafael Wysocki)

   - Add support for Clearwater Forest in the out-of-band (OOB) mode to
     the intel_pstate driver (Srinivas Pandruvada)"

* tag 'pm-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  cpufreq: intel_pstate: Support Clearwater Forest OOB mode
  cpuidle: governors: menu: Avoid using invalid recent intervals data
  intel_idle: Allow loading ACPI tables for any family

2 months agodrm/xe/pf: Set VF LMEM BAR size
Michał Winiarski [Tue, 27 May 2025 12:06:37 +0000 (14:06 +0200)]
drm/xe/pf: Set VF LMEM BAR size

LMEM is partitioned between multiple VFs and we expect that the more
VFs we have, the less LMEM is assigned to each VF.
This means that we can achieve full LMEM BAR access without the need to
attempt full VF LMEM BAR resize via pci_resize_resource().

Always try to set the largest possible BAR size that allows to fit the
number of enabled VFs and inform the user in case the resize attempt is
not successful.

Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20250527120637.665506-7-michal.winiarski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit 32a4d1b98e6663101fd0abfaf151c48feea7abb1)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
2 months agoMerge tag 'net-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Thu, 14 Aug 2025 14:14:30 +0000 (07:14 -0700)]
Merge tag 'net-6.17-rc2' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Paolo Abeni:
 "Including fixes from Netfilter and IPsec.

  Current release - regressions:

   - netfilter: nft_set_pipapo:
      - don't return bogus extension pointer
      - fix null deref for empty set

  Current release - new code bugs:

   - core: prevent deadlocks when enabling NAPIs with mixed kthread
     config

   - eth: netdevsim: Fix wild pointer access in nsim_queue_free().

  Previous releases - regressions:

   - page_pool: allow enabling recycling late, fix false positive
     warning

   - sched: ets: use old 'nbands' while purging unused classes

   - xfrm:
      - restore GSO for SW crypto
      - bring back device check in validate_xmit_xfrm

   - tls: handle data disappearing from under the TLS ULP

   - ptp: prevent possible ABBA deadlock in ptp_clock_freerun()

   - eth:
      - bnxt: fill data page pool with frags if PAGE_SIZE > BNXT_RX_PAGE_SIZE
      - hv_netvsc: fix panic during namespace deletion with VF

  Previous releases - always broken:

   - netfilter: fix refcount leak on table dump

   - vsock: do not allow binding to VMADDR_PORT_ANY

   - sctp: linearize cloned gso packets in sctp_rcv

   - eth:
      - hibmcge: fix the division by zero issue
      - microchip: fix KSZ8863 reset problem"

* tag 'net-6.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (54 commits)
  net: usb: asix_devices: add phy_mask for ax88772 mdio bus
  net: kcm: Fix race condition in kcm_unattach()
  selftests: net/forwarding: test purge of active DWRR classes
  net/sched: ets: use old 'nbands' while purging unused classes
  bnxt: fill data page pool with frags if PAGE_SIZE > BNXT_RX_PAGE_SIZE
  netdevsim: Fix wild pointer access in nsim_queue_free().
  net: mctp: Fix bad kfree_skb in bind lookup test
  netfilter: nf_tables: reject duplicate device on updates
  ipvs: Fix estimator kthreads preferred affinity
  netfilter: nft_set_pipapo: fix null deref for empty set
  selftests: tls: test TCP stealing data from under the TLS socket
  tls: handle data disappearing from under the TLS ULP
  ptp: prevent possible ABBA deadlock in ptp_clock_freerun()
  ixgbe: prevent from unwanted interface name changes
  devlink: let driver opt out of automatic phys_port_name generation
  net: prevent deadlocks when enabling NAPIs with mixed kthread config
  net: update NAPI threaded config even for disabled NAPIs
  selftests: drv-net: don't assume device has only 2 queues
  docs: Fix name for net.ipv4.udp_child_hash_entries
  riscv: dts: thead: Add APB clocks for TH1520 GMACs
  ...

2 months agoMerge branches 'acpi-ec' and 'acpi-processor'
Rafael J. Wysocki [Thu, 14 Aug 2025 11:09:27 +0000 (13:09 +0200)]
Merge branches 'acpi-ec' and 'acpi-processor'

* acpi-ec:
  ACPI: EC: Relax sanity check of the ECDT ID string

* acpi-processor:
  ACPI: processor: perflib: Move problematic pr->performance check

2 months agoMerge branches 'pm-cpuidle' and 'pm-cpufreq'
Rafael J. Wysocki [Thu, 14 Aug 2025 10:57:11 +0000 (12:57 +0200)]
Merge branches 'pm-cpuidle' and 'pm-cpufreq'

* pm-cpuidle:
  cpuidle: governors: menu: Avoid using invalid recent intervals data
  intel_idle: Allow loading ACPI tables for any family

* pm-cpufreq:
  cpufreq: intel_pstate: Support Clearwater Forest OOB mode

2 months agoata: libata-scsi: Fix CDL control
Igor Pylypiv [Thu, 14 Aug 2025 02:22:56 +0000 (19:22 -0700)]
ata: libata-scsi: Fix CDL control

Delete extra checks for the ATA_DFLAG_CDL_ENABLED flag that prevent
SET FEATURES command from being issued to a drive when NCQ commands
are active.

ata_mselect_control_ata_feature() sets / clears the ATA_DFLAG_CDL_ENABLED
flag during the translation of MODE SELECT to SET FEATURES. If SET FEATURES
gets deferred due to outstanding NCQ commands, the original MODE SELECT
command will be re-queued. When the re-queued MODE SELECT goes through
the ata_mselect_control_ata_feature() translation again, SET FEATURES
will not be issued because ATA_DFLAG_CDL_ENABLED has been already set or
cleared by the initial translation of MODE SELECT.

The ATA_DFLAG_CDL_ENABLED checks in ata_mselect_control_ata_feature()
are safe to remove because scsi_cdl_enable() implements a similar logic
that avoids enabling CDL if it has been enabled already.

Fixes: 17e897a45675 ("ata: libata-scsi: Improve CDL control")
Cc: stable@vger.kernel.org
Signed-off-by: Igor Pylypiv <ipylypiv@google.com>
Reviewed-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
2 months agoata: libata-eh: Fix link state check for IDE/PATA ports
Damien Le Moal [Wed, 13 Aug 2025 09:08:53 +0000 (18:08 +0900)]
ata: libata-eh: Fix link state check for IDE/PATA ports

Commit 4371fe1ba400 ("ata: libata-eh: Avoid unnecessary resets when
revalidating devices") replaced the call to ata_phys_link_offline() in
ata_eh_revalidate_and_attach() with the new function
ata_eh_link_established() which relaxes the checks on a device link
state to account for low power mode transitions. However, this change
assumed that the device port has a valid scr_read method to obtain the
SStatus register for the port. This is not always the case, especially
with older IDE/PATA adapters (e.g. PATA/IDE devices emulated with QEMU).
For such adapter, ata_eh_link_established() will always return false,
causing ata_eh_revalidate_and_attach() to go into its error path and
ultimately to the device being disabled.

Avoid this by restoring the previous behavior, which is to assume that
the link is online if reading the port SStatus register fails.

While at it, also fix the spelling of SStatus in the comment describing
the function ata_eh_link_established().

Reported-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Fixes: 4371fe1ba400 ("ata: libata-eh: Avoid unnecessary resets when revalidating devices")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reviewed-by: Niklas Cassel <cassel@kernel.org>
2 months agoALSA: hda/tas2781: Normalize the volume kcontrol name
Baojun Xu [Wed, 13 Aug 2025 10:08:42 +0000 (18:08 +0800)]
ALSA: hda/tas2781: Normalize the volume kcontrol name

Change the name of the kcontrol from "Gain" to "Volume".

Signed-off-by: Baojun Xu <baojun.xu@ti.com>
Link: https://patch.msgid.link/20250813100842.12224-1-baojun.xu@ti.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2 months agoALSA: usb-audio: Validate UAC3 cluster segment descriptors
Takashi Iwai [Thu, 14 Aug 2025 08:12:43 +0000 (10:12 +0200)]
ALSA: usb-audio: Validate UAC3 cluster segment descriptors

UAC3 class segment descriptors need to be verified whether their sizes
match with the declared lengths and whether they fit with the
allocated buffer sizes, too.  Otherwise malicious firmware may lead to
the unexpected OOB accesses.

Fixes: 11785ef53228 ("ALSA: usb-audio: Initial Power Domain support")
Reported-and-tested-by: Youngjun Lee <yjjuny.lee@samsung.com>
Cc: <stable@vger.kernel.org>
Link: https://patch.msgid.link/20250814081245.8902-2-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2 months agoALSA: usb-audio: Validate UAC3 power domain descriptors, too
Takashi Iwai [Thu, 14 Aug 2025 08:12:42 +0000 (10:12 +0200)]
ALSA: usb-audio: Validate UAC3 power domain descriptors, too

UAC3 power domain descriptors need to be verified with its variable
bLength for avoiding the unexpected OOB accesses by malicious
firmware, too.

Fixes: 9a2fe9b801f5 ("ALSA: usb: initial USB Audio Device Class 3.0 support")
Reported-and-tested-by: Youngjun Lee <yjjuny.lee@samsung.com>
Cc: <stable@vger.kernel.org>
Link: https://patch.msgid.link/20250814081245.8902-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2 months agonet: usb: asix_devices: add phy_mask for ax88772 mdio bus
Xu Yang [Mon, 11 Aug 2025 09:29:31 +0000 (17:29 +0800)]
net: usb: asix_devices: add phy_mask for ax88772 mdio bus

Without setting phy_mask for ax88772 mdio bus, current driver may create
at most 32 mdio phy devices with phy address range from 0x00 ~ 0x1f.
DLink DUB-E100 H/W Ver B1 is such a device. However, only one main phy
device will bind to net phy driver. This is creating issue during system
suspend/resume since phy_polling_mode() in phy_state_machine() will
directly deference member of phydev->drv for non-main phy devices. Then
NULL pointer dereference issue will occur. Due to only external phy or
internal phy is necessary, add phy_mask for ax88772 mdio bus to workarnoud
the issue.

Closes: https://lore.kernel.org/netdev/20250806082931.3289134-1-xu.yang_2@nxp.com
Fixes: e532a096be0e ("net: usb: asix: ax88772: add phylib support")
Cc: stable@vger.kernel.org
Signed-off-by: Xu Yang <xu.yang_2@nxp.com>
Tested-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://patch.msgid.link/20250811092931.860333-1-xu.yang_2@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 months agoMerge tag 'asoc-fix-v6.17-rc1' of https://git.kernel.org/pub/scm/linux/kernel/git...
Takashi Iwai [Thu, 14 Aug 2025 06:33:44 +0000 (08:33 +0200)]
Merge tag 'asoc-fix-v6.17-rc1' of https://git./linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v6.17

A reasonably small collection of fixes that came in since the merge
window, mostly small and driver specific plus a cleanup of the menu
reorganisation to address some user confusion with the way the generic
drivers had been handled.

2 months agoMerge tag 'probes-fixes-v6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 14 Aug 2025 03:23:32 +0000 (20:23 -0700)]
Merge tag 'probes-fixes-v6.17-rc1' of git://git./linux/kernel/git/trace/linux-trace

Pull probes fix from Masami Hiramatsu:

 - MAINTAINERS: Remove bouncing kprobes maintainer

* tag 'probes-fixes-v6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  MAINTAINERS: Remove bouncing kprobes maintainer

2 months agoMAINTAINERS: Remove bouncing kprobes maintainer
Dave Hansen [Thu, 14 Aug 2025 02:38:58 +0000 (11:38 +0900)]
MAINTAINERS: Remove bouncing kprobes maintainer

The kprobes MAINTAINERS entry includes anil.s.keshavamurthy@intel.com.
That address is bouncing. Remove it.

This still leaves three other listed maintainers.

Link: https://lore.kernel.org/all/20250808180124.7DDE2ECD@davehans-spike.ostc.intel.com/
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Naveen N Rao <naveen@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: linux-trace-kernel@vger.kernel.org
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
2 months agonet: kcm: Fix race condition in kcm_unattach()
Sven Stegemann [Tue, 12 Aug 2025 19:18:03 +0000 (21:18 +0200)]
net: kcm: Fix race condition in kcm_unattach()

syzbot found a race condition when kcm_unattach(psock)
and kcm_release(kcm) are executed at the same time.

kcm_unattach() is missing a check of the flag
kcm->tx_stopped before calling queue_work().

If the kcm has a reserved psock, kcm_unattach() might get executed
between cancel_work_sync() and unreserve_psock() in kcm_release(),
requeuing kcm->tx_work right before kcm gets freed in kcm_done().

Remove kcm->tx_stopped and replace it by the less
error-prone disable_work_sync().

Fixes: ab7ac4eb9832 ("kcm: Kernel Connection Multiplexor module")
Reported-by: syzbot+e62c9db591c30e174662@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=e62c9db591c30e174662
Reported-by: syzbot+d199b52665b6c3069b94@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=d199b52665b6c3069b94
Reported-by: syzbot+be6b1fdfeae512726b4e@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=be6b1fdfeae512726b4e
Signed-off-by: Sven Stegemann <sven@stegemann.de>
Link: https://patch.msgid.link/20250812191810.27777-1-sven@stegemann.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agoMerge branch 'ets-use-old-nbands-while-purging-unused-classes'
Jakub Kicinski [Thu, 14 Aug 2025 01:11:56 +0000 (18:11 -0700)]
Merge branch 'ets-use-old-nbands-while-purging-unused-classes'

Davide Caratti says:

====================
ets: use old 'nbands' while purging unused classes

- patch 1/2 fixes a NULL dereference in the control path of sch_ets qdisc
- patch 2/2 extends kselftests to verify effectiveness of the above fix
====================

Link: https://patch.msgid.link/cover.1755016081.git.dcaratti@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agoselftests: net/forwarding: test purge of active DWRR classes
Davide Caratti [Tue, 12 Aug 2025 16:40:30 +0000 (18:40 +0200)]
selftests: net/forwarding: test purge of active DWRR classes

Extend sch_ets.sh to add a reproducer for problematic list deletions when
active DWRR class are purged by ets_qdisc_change() [1] [2].

[1] https://lore.kernel.org/netdev/e08c7f4a6882f260011909a868311c6e9b54f3e4.1639153474.git.dcaratti@redhat.com/
[2] https://lore.kernel.org/netdev/f3b9bacc73145f265c19ab80785933da5b7cbdec.1754581577.git.dcaratti@redhat.com/

Suggested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Acked-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/489497cb781af7389011ca1591fb702a7391f5e7.1755016081.git.dcaratti@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agonet/sched: ets: use old 'nbands' while purging unused classes
Davide Caratti [Tue, 12 Aug 2025 16:40:29 +0000 (18:40 +0200)]
net/sched: ets: use old 'nbands' while purging unused classes

Shuang reported sch_ets test-case [1] crashing in ets_class_qlen_notify()
after recent changes from Lion [2]. The problem is: in ets_qdisc_change()
we purge unused DWRR queues; the value of 'q->nbands' is the new one, and
the cleanup should be done with the old one. The problem is here since my
first attempts to fix ets_qdisc_change(), but it surfaced again after the
recent qdisc len accounting fixes. Fix it purging idle DWRR queues before
assigning a new value of 'q->nbands', so that all purge operations find a
consistent configuration:

 - old 'q->nbands' because it's needed by ets_class_find()
 - old 'q->nstrict' because it's needed by ets_class_is_strict()

 BUG: kernel NULL pointer dereference, address: 0000000000000000
 #PF: supervisor read access in kernel mode
 #PF: error_code(0x0000) - not-present page
 PGD 0 P4D 0
 Oops: Oops: 0000 [#1] SMP NOPTI
 CPU: 62 UID: 0 PID: 39457 Comm: tc Kdump: loaded Not tainted 6.12.0-116.el10.x86_64 #1 PREEMPT(voluntary)
 Hardware name: Dell Inc. PowerEdge R640/06DKY5, BIOS 2.12.2 07/09/2021
 RIP: 0010:__list_del_entry_valid_or_report+0x4/0x80
 Code: ff 4c 39 c7 0f 84 39 19 8e ff b8 01 00 00 00 c3 cc cc cc cc 66 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 f3 0f 1e fa <48> 8b 17 48 8b 4f 08 48 85 d2 0f 84 56 19 8e ff 48 85 c9 0f 84 ab
 RSP: 0018:ffffba186009f400 EFLAGS: 00010202
 RAX: 00000000000000d6 RBX: 0000000000000000 RCX: 0000000000000004
 RDX: ffff9f0fa29b69c0 RSI: 0000000000000000 RDI: 0000000000000000
 RBP: ffffffffc12c2400 R08: 0000000000000008 R09: 0000000000000004
 R10: ffffffffffffffff R11: 0000000000000004 R12: 0000000000000000
 R13: ffff9f0f8cfe0000 R14: 0000000000100005 R15: 0000000000000000
 FS:  00007f2154f37480(0000) GS:ffff9f269c1c0000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 0000000000000000 CR3: 00000001530be001 CR4: 00000000007726f0
 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
 PKRU: 55555554
 Call Trace:
  <TASK>
  ets_class_qlen_notify+0x65/0x90 [sch_ets]
  qdisc_tree_reduce_backlog+0x74/0x110
  ets_qdisc_change+0x630/0xa40 [sch_ets]
  __tc_modify_qdisc.constprop.0+0x216/0x7f0
  tc_modify_qdisc+0x7c/0x120
  rtnetlink_rcv_msg+0x145/0x3f0
  netlink_rcv_skb+0x53/0x100
  netlink_unicast+0x245/0x390
  netlink_sendmsg+0x21b/0x470
  ____sys_sendmsg+0x39d/0x3d0
  ___sys_sendmsg+0x9a/0xe0
  __sys_sendmsg+0x7a/0xd0
  do_syscall_64+0x7d/0x160
  entry_SYSCALL_64_after_hwframe+0x76/0x7e
 RIP: 0033:0x7f2155114084
 Code: 89 02 b8 ff ff ff ff eb bb 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 80 3d 25 f0 0c 00 00 74 13 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 c3 0f 1f 00 48 83 ec 28 89 54 24 1c 48 89
 RSP: 002b:00007fff1fd7a988 EFLAGS: 00000202 ORIG_RAX: 000000000000002e
 RAX: ffffffffffffffda RBX: 0000560ec063e5e0 RCX: 00007f2155114084
 RDX: 0000000000000000 RSI: 00007fff1fd7a9f0 RDI: 0000000000000003
 RBP: 00007fff1fd7aa60 R08: 0000000000000010 R09: 000000000000003f
 R10: 0000560ee9b3a010 R11: 0000000000000202 R12: 00007fff1fd7aae0
 R13: 000000006891ccde R14: 0000560ec063e5e0 R15: 00007fff1fd7aad0
  </TASK>

 [1] https://lore.kernel.org/netdev/e08c7f4a6882f260011909a868311c6e9b54f3e4.1639153474.git.dcaratti@redhat.com/
 [2] https://lore.kernel.org/netdev/d912cbd7-193b-4269-9857-525bee8bbb6a@gmail.com/

Cc: stable@vger.kernel.org
Fixes: 103406b38c60 ("net/sched: Always pass notifications when child class becomes empty")
Fixes: c062f2a0b04d ("net/sched: sch_ets: don't remove idle classes from the round-robin list")
Fixes: dcc68b4d8084 ("net: sch_ets: Add a new Qdisc")
Reported-by: Li Shuang <shuali@redhat.com>
Closes: https://issues.redhat.com/browse/RHEL-108026
Reviewed-by: Petr Machata <petrm@nvidia.com>
Co-developed-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/7928ff6d17db47a2ae7cc205c44777b1f1950545.1755016081.git.dcaratti@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agoMerge branch '10GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Jakub Kicinski [Thu, 14 Aug 2025 00:31:46 +0000 (17:31 -0700)]
Merge branch '10GbE' of git://git./linux/kernel/git/tnguy/net-queue

Tony Nguyen says:

====================
ixgbe: bypass devlink phys_port_name generation

Jedrzej adds option to skip phys_port_name generation and opts
ixgbe into it as some configurations rely on pre-devlink naming
which could end up broken as a result.

* '10GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
  ixgbe: prevent from unwanted interface name changes
  devlink: let driver opt out of automatic phys_port_name generation
====================

Link: https://patch.msgid.link/20250812205226.1984369-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agobnxt: fill data page pool with frags if PAGE_SIZE > BNXT_RX_PAGE_SIZE
David Wei [Tue, 12 Aug 2025 18:29:07 +0000 (11:29 -0700)]
bnxt: fill data page pool with frags if PAGE_SIZE > BNXT_RX_PAGE_SIZE

The data page pool always fills the HW rx ring with pages. On arm64 with
64K pages, this will waste _at least_ 32K of memory per entry in the rx
ring.

Fix by fragmenting the pages if PAGE_SIZE > BNXT_RX_PAGE_SIZE. This
makes the data page pool the same as the header pool.

Tested with iperf3 with a small (64 entries) rx ring to encourage buffer
circulation.

Fixes: cd1fafe7da1f ("eth: bnxt: add support rx side device memory TCP")
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David Wei <dw@davidwei.uk>
Link: https://patch.msgid.link/20250812182907.1540755-1-dw@davidwei.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agonetdevsim: Fix wild pointer access in nsim_queue_free().
Kuniyuki Iwashima [Tue, 12 Aug 2025 16:21:26 +0000 (16:21 +0000)]
netdevsim: Fix wild pointer access in nsim_queue_free().

syzbot reported the splat below. [0]

When nsim_queue_uninit() is called from nsim_init_netdevsim(),
register_netdevice() has not been called, thus dev->dstats has
not been allocated.

Let's not call dev_dstats_rx_dropped_add() in such a case.

[0]
BUG: unable to handle page fault for address: ffff88809782c020
 PF: supervisor write access in kernel mode
 PF: error_code(0x0002) - not-present page
PGD 1b401067 P4D 1b401067 PUD 0
Oops: Oops: 0002 [#1] SMP KASAN NOPTI
CPU: 3 UID: 0 PID: 8476 Comm: syz.1.251 Not tainted 6.16.0-syzkaller-06699-ge8d780dcd957 #0 PREEMPT(full)
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
RIP: 0010:local_add arch/x86/include/asm/local.h:33 [inline]
RIP: 0010:u64_stats_add include/linux/u64_stats_sync.h:89 [inline]
RIP: 0010:dev_dstats_rx_dropped_add include/linux/netdevice.h:3027 [inline]
RIP: 0010:nsim_queue_free+0xba/0x120 drivers/net/netdevsim/netdev.c:714
Code: 07 77 6c 4a 8d 3c ed 20 7e f1 8d 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 80 3c 02 00 75 46 4a 03 1c ed 20 7e f1 8d <4c> 01 63 20 be 00 02 00 00 48 8d 3d 00 00 00 00 e8 61 2f 58 fa 48
RSP: 0018:ffffc900044af150 EFLAGS: 00010286
RAX: dffffc0000000000 RBX: ffff88809782c000 RCX: 00000000000079c3
RDX: 1ffffffff1be2fc7 RSI: ffffffff8c15f380 RDI: ffffffff8df17e38
RBP: ffff88805f59d000 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000001 R12: 0000000000000000
R13: 0000000000000003 R14: ffff88806ceb3d00 R15: ffffed100dfd308e
FS:  0000000000000000(0000) GS:ffff88809782c000(0063) knlGS:00000000f505db40
CS:  0010 DS: 002b ES: 002b CR0: 0000000080050033
CR2: ffff88809782c020 CR3: 000000006fc6a000 CR4: 0000000000352ef0
Call Trace:
 <TASK>
 nsim_queue_uninit drivers/net/netdevsim/netdev.c:993 [inline]
 nsim_init_netdevsim drivers/net/netdevsim/netdev.c:1049 [inline]
 nsim_create+0xd0a/0x1260 drivers/net/netdevsim/netdev.c:1101
 __nsim_dev_port_add+0x435/0x7d0 drivers/net/netdevsim/dev.c:1438
 nsim_dev_port_add_all drivers/net/netdevsim/dev.c:1494 [inline]
 nsim_dev_reload_create drivers/net/netdevsim/dev.c:1546 [inline]
 nsim_dev_reload_up+0x5b8/0x860 drivers/net/netdevsim/dev.c:1003
 devlink_reload+0x322/0x7c0 net/devlink/dev.c:474
 devlink_nl_reload_doit+0xe31/0x1410 net/devlink/dev.c:584
 genl_family_rcv_msg_doit+0x206/0x2f0 net/netlink/genetlink.c:1115
 genl_family_rcv_msg net/netlink/genetlink.c:1195 [inline]
 genl_rcv_msg+0x55c/0x800 net/netlink/genetlink.c:1210
 netlink_rcv_skb+0x155/0x420 net/netlink/af_netlink.c:2552
 genl_rcv+0x28/0x40 net/netlink/genetlink.c:1219
 netlink_unicast_kernel net/netlink/af_netlink.c:1320 [inline]
 netlink_unicast+0x5aa/0x870 net/netlink/af_netlink.c:1346
 netlink_sendmsg+0x8d1/0xdd0 net/netlink/af_netlink.c:1896
 sock_sendmsg_nosec net/socket.c:714 [inline]
 __sock_sendmsg net/socket.c:729 [inline]
 ____sys_sendmsg+0xa95/0xc70 net/socket.c:2614
 ___sys_sendmsg+0x134/0x1d0 net/socket.c:2668
 __sys_sendmsg+0x16d/0x220 net/socket.c:2700
 do_syscall_32_irqs_on arch/x86/entry/syscall_32.c:83 [inline]
 __do_fast_syscall_32+0x7c/0x3a0 arch/x86/entry/syscall_32.c:306
 do_fast_syscall_32+0x32/0x80 arch/x86/entry/syscall_32.c:331
 entry_SYSENTER_compat_after_hwframe+0x84/0x8e
RIP: 0023:0xf708e579
Code: b8 01 10 06 03 74 b4 01 10 07 03 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 00 00 00 00 00 00 00 00 51 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d b4 26 00 00 00 00 8d b4 26 00 00 00 00
RSP: 002b:00000000f505d55c EFLAGS: 00000296 ORIG_RAX: 0000000000000172
RAX: ffffffffffffffda RBX: 0000000000000007 RCX: 0000000080000080
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000296 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
 </TASK>
Modules linked in:
CR2: ffff88809782c020

Fixes: 2a68a22304f9 ("netdevsim: account dropped packet length in stats on queue free")
Reported-by: syzbot+8aa80c6232008f7b957d@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/688bb9ca.a00a0220.26d0e1.0050.GAE@google.com/
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20250812162130.4129322-1-kuniyu@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agonet: mctp: Fix bad kfree_skb in bind lookup test
Matt Johnston [Tue, 12 Aug 2025 05:08:58 +0000 (13:08 +0800)]
net: mctp: Fix bad kfree_skb in bind lookup test

The kunit test's skb_pkt is consumed by mctp_dst_input() so shouldn't be
freed separately.

Fixes: e6d8e7dbc5a3 ("net: mctp: Add bind lookup test")
Reported-by: Alexandre Ghiti <alex@ghiti.fr>
Closes: https://lore.kernel.org/all/734b02a3-1941-49df-a0da-ec14310d41e4@ghiti.fr/
Signed-off-by: Matt Johnston <matt@codeconstruct.com.au>
Tested-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Link: https://patch.msgid.link/20250812-fix-mctp-bind-test-v1-1-5e2128664eb3@codeconstruct.com.au
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agoMerge tag 'amd-drm-fixes-6.17-2025-08-13' of https://gitlab.freedesktop.org/agd5f...
Dave Airlie [Wed, 13 Aug 2025 22:16:25 +0000 (08:16 +1000)]
Merge tag 'amd-drm-fixes-6.17-2025-08-13' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.17-2025-08-13:

amdgpu:
- PSP fix
- VRAM reservation fix
- CSA fix
- Process kill fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250813151905.2040816-1-alexander.deucher@amd.com
2 months agoMerge tag 'nf-25-08-13' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Jakub Kicinski [Wed, 13 Aug 2025 21:51:51 +0000 (14:51 -0700)]
Merge tag 'nf-25-08-13' of https://git./linux/kernel/git/netfilter/nf

Florian Westphal says:

====================
Netfilter fixes for net

The following patchset contains Netfilter fixes for *net*:

1) I managed to add a null dereference crash in nft_set_pipapo
   in the current development cycle, was not caught by CI
   because the avx2 implementation is fine, but selftest
   splats when run on non-avx2 host.

2) Fix the ipvs estimater kthread affinity, was incorrect
   since 6.14. From Frederic Weisbecker.

3) nf_tables should not allow to add a device to a flowtable
   or netdev chain more than once -- reject this.
   From Pablo Neira Ayuso.  This has been broken for long time,
   blamed commit dates from v5.8.

* tag 'nf-25-08-13' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
  netfilter: nf_tables: reject duplicate device on updates
  ipvs: Fix estimator kthreads preferred affinity
  netfilter: nft_set_pipapo: fix null deref for empty set
====================

Link: https://patch.msgid.link/20250813113800.20775-1-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 months agoMerge tag 'drm-misc-next-fixes-2025-08-12' of https://gitlab.freedesktop.org/drm...
Dave Airlie [Wed, 13 Aug 2025 21:50:40 +0000 (07:50 +1000)]
Merge tag 'drm-misc-next-fixes-2025-08-12' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes

Short summary of fixes pull:

bridge:
- fix OF-node leak
- fix documentation

fbdev-emulation:
- pass correct format info to drm_helper_mode_fill_fb_struct()

panfrost:
- print correct RSS size

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250812064712.GA14554@2a02-2454-fd5e-fd00-2c49-c639-c55f-a125.dyn6.pyur.net
2 months agoMerge tag 'erofs-for-6.17-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 13 Aug 2025 18:29:27 +0000 (11:29 -0700)]
Merge tag 'erofs-for-6.17-rc2-fixes' of git://git./linux/kernel/git/xiang/erofs

Pull erofs fixes from Gao Xiang:

 - Align FSDAX enablement among multiple devices

 - Fix EROFS_FS_ZIP_ACCEL build dependency again to prevent forcing
   CRYPTO{,_DEFLATE}=y even if EROFS=m

 - Fix atomic context detection to properly launch kworkers on demand

 - Fix block count statistics for 48-bit addressing support

* tag 'erofs-for-6.17-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
  erofs: fix block count report when 48-bit layout is on
  erofs: fix atomic context detection when !CONFIG_DEBUG_LOCK_ALLOC
  erofs: Do not select tristate symbols from bool symbols
  erofs: Fallback to normal access if DAX is not supported on extra device

2 months agoMerge tag 'rcu.fixes.6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux
Linus Torvalds [Wed, 13 Aug 2025 17:23:28 +0000 (10:23 -0700)]
Merge tag 'rcu.fixes.6.17' of git://git./linux/kernel/git/rcu/linux

Pull RCU fix from Neeraj Upadhyay:
 "Fix a regression introduced by commit b41642c87716 ("rcu: Fix
  rcu_read_unlock() deadloop due to IRQ work") which results in boot
  hang as reported by kernel test bot at [1].

  This issue happens because RCU re-initializes the deferred QS IRQ work
  everytime it is queued. With commit b41642c87716, the IRQ work
  re-initialization can happen while it is already queued. This results
  in IRQ work being requeued to itself. When IRQ work finally fires, as
  it is requeued to itself, it is repeatedly executed and results in
  hang.

  Fix this with initializing the IRQ work only once before the CPU
  boots"

Link: https://lore.kernel.org/rcu/202508071303.c1134cce-lkp@intel.com/
* tag 'rcu.fixes.6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux:
  rcu: Fix racy re-initialization of irq_work causing hangs

2 months agosmb: client: remove redundant lstrp update in negotiate protocol
Wang Zhaolong [Fri, 1 Aug 2025 09:07:24 +0000 (17:07 +0800)]
smb: client: remove redundant lstrp update in negotiate protocol

Commit 34331d7beed7 ("smb: client: fix first command failure during
re-negotiation") addressed a race condition by updating lstrp before
entering negotiate state. However, this approach may have some unintended
side effects.

The lstrp field is documented as "when we got last response from this
server", and updating it before actually receiving a server response
could potentially affect other mechanisms that rely on this timestamp.
For example, the SMB echo detection logic also uses lstrp as a reference
point. In scenarios with frequent user operations during reconnect states,
the repeated calls to cifs_negotiate_protocol() might continuously
update lstrp, which could interfere with the echo detection timing.

Additionally, commit 266b5d02e14f ("smb: client: fix race condition in
negotiate timeout by using more precise timing") introduced a dedicated
neg_start field specifically for tracking negotiate start time. This
provides a more precise solution for the original race condition while
preserving the intended semantics of lstrp.

Since the race condition is now properly handled by the neg_start
mechanism, the lstrp update in cifs_negotiate_protocol() is no longer
necessary and can be safely removed.

Fixes: 266b5d02e14f ("smb: client: fix race condition in negotiate timeout by using more precise timing")
Cc: stable@vger.kernel.org
Acked-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Signed-off-by: Wang Zhaolong <wangzhaolong@huaweicloud.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
2 months agocifs: update internal version number
Steve French [Sat, 9 Aug 2025 14:17:46 +0000 (09:17 -0500)]
cifs: update internal version number

to 2.56

Signed-off-by: Steve French <stfrench@microsoft.com>
2 months agosmb: client: don't wait for info->send_pending == 0 on error
Stefan Metzmacher [Tue, 12 Aug 2025 16:45:06 +0000 (18:45 +0200)]
smb: client: don't wait for info->send_pending == 0 on error

We already called ib_drain_qp() before and that makes sure
send_done() was called with IB_WC_WR_FLUSH_ERR, but
didn't called atomic_dec_and_test(&sc->send_io.pending.count)

So we may never reach the info->send_pending == 0 condition.

Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Fixes: 5349ae5e05fa ("smb: client: let send_done() cleanup before calling smbd_disconnect_rdma_connection()")
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
2 months agosmb: client: fix mid_q_entry memleak leak with per-mid locking
Wang Zhaolong [Mon, 11 Aug 2025 14:07:37 +0000 (22:07 +0800)]
smb: client: fix mid_q_entry memleak leak with per-mid locking

This is step 4/4 of a patch series to fix mid_q_entry memory leaks
caused by race conditions in callback execution.

In compound_send_recv(), when wait_for_response() is interrupted by
signals, the code attempts to cancel pending requests by changing
their callbacks to cifs_cancelled_callback. However, there's a race
condition between signal interruption and network response processing
that causes both mid_q_entry and server buffer leaks:

```
User foreground process                    cifsd
cifs_readdir
 open_cached_dir
  cifs_send_recv
   compound_send_recv
    smb2_setup_request
     smb2_mid_entry_alloc
      smb2_get_mid_entry
       smb2_mid_entry_alloc
        mempool_alloc // alloc mid
        kref_init(&temp->refcount); // refcount = 1
     mid[0]->callback = cifs_compound_callback;
     mid[1]->callback = cifs_compound_last_callback;
     smb_send_rqst
     rc = wait_for_response
      wait_event_state TASK_KILLABLE
                                  cifs_demultiplex_thread
                                    allocate_buffers
                                      server->bigbuf = cifs_buf_get()
                                    standard_receive3
                                      ->find_mid()
                                        smb2_find_mid
                                          __smb2_find_mid
                                           kref_get(&mid->refcount) // +1
                                      cifs_handle_standard
                                        handle_mid
                                         /* bigbuf will also leak */
                                         mid->resp_buf = server->bigbuf
                                         server->bigbuf = NULL;
                                         dequeue_mid
                                     /* in for loop */
                                    mids[0]->callback
                                      cifs_compound_callback
    /* Signal interrupts wait: rc = -ERESTARTSYS */
    /* if (... || midQ[i]->mid_state == MID_RESPONSE_RECEIVED) *?
    midQ[0]->callback = cifs_cancelled_callback;
    cancelled_mid[i] = true;
                                       /* The change comes too late */
                                       mid->mid_state = MID_RESPONSE_READY
                                    release_mid  // -1
    /* cancelled_mid[i] == true causes mid won't be released
       in compound_send_recv cleanup */
    /* cifs_cancelled_callback won't executed to release mid */
```

The root cause is that there's a race between callback assignment and
execution.

Fix this by introducing per-mid locking:

- Add spinlock_t mid_lock to struct mid_q_entry
- Add mid_execute_callback() for atomic callback execution
- Use mid_lock in cancellation paths to ensure atomicity

This ensures that either the original callback or the cancellation
callback executes atomically, preventing reference count leaks when
requests are interrupted by signals.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=220404
Fixes: ee258d79159a ("CIFS: Move credit processing to mid callbacks for SMB3")
Signed-off-by: Wang Zhaolong <wangzhaolong@huaweicloud.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
2 months agosmb3: fix for slab out of bounds on mount to ksmbd
Steve French [Tue, 12 Aug 2025 04:14:55 +0000 (23:14 -0500)]
smb3: fix for slab out of bounds on mount to ksmbd

With KASAN enabled, it is possible to get a slab out of bounds
during mount to ksmbd due to missing check in parse_server_interfaces()
(see below):

 BUG: KASAN: slab-out-of-bounds in
 parse_server_interfaces+0x14ee/0x1880 [cifs]
 Read of size 4 at addr ffff8881433dba98 by task mount/9827

 CPU: 5 UID: 0 PID: 9827 Comm: mount Tainted: G
 OE       6.16.0-rc2-kasan #2 PREEMPT(voluntary)
 Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
 Hardware name: Dell Inc. Precision Tower 3620/0MWYPT,
 BIOS 2.13.1 06/14/2019
 Call Trace:
  <TASK>
 dump_stack_lvl+0x9f/0xf0
 print_report+0xd1/0x670
 __virt_addr_valid+0x22c/0x430
 ? parse_server_interfaces+0x14ee/0x1880 [cifs]
 ? kasan_complete_mode_report_info+0x2a/0x1f0
 ? parse_server_interfaces+0x14ee/0x1880 [cifs]
   kasan_report+0xd6/0x110
   parse_server_interfaces+0x14ee/0x1880 [cifs]
   __asan_report_load_n_noabort+0x13/0x20
   parse_server_interfaces+0x14ee/0x1880 [cifs]
 ? __pfx_parse_server_interfaces+0x10/0x10 [cifs]
 ? trace_hardirqs_on+0x51/0x60
 SMB3_request_interfaces+0x1ad/0x3f0 [cifs]
 ? __pfx_SMB3_request_interfaces+0x10/0x10 [cifs]
 ? SMB2_tcon+0x23c/0x15d0 [cifs]
 smb3_qfs_tcon+0x173/0x2b0 [cifs]
 ? __pfx_smb3_qfs_tcon+0x10/0x10 [cifs]
 ? cifs_get_tcon+0x105d/0x2120 [cifs]
 ? do_raw_spin_unlock+0x5d/0x200
 ? cifs_get_tcon+0x105d/0x2120 [cifs]
 ? __pfx_smb3_qfs_tcon+0x10/0x10 [cifs]
 cifs_mount_get_tcon+0x369/0xb90 [cifs]
 ? dfs_cache_find+0xe7/0x150 [cifs]
 dfs_mount_share+0x985/0x2970 [cifs]
 ? check_path.constprop.0+0x28/0x50
 ? save_trace+0x54/0x370
 ? __pfx_dfs_mount_share+0x10/0x10 [cifs]
 ? __lock_acquire+0xb82/0x2ba0
 ? __kasan_check_write+0x18/0x20
 cifs_mount+0xbc/0x9e0 [cifs]
 ? __pfx_cifs_mount+0x10/0x10 [cifs]
 ? do_raw_spin_unlock+0x5d/0x200
 ? cifs_setup_cifs_sb+0x29d/0x810 [cifs]
 cifs_smb3_do_mount+0x263/0x1990 [cifs]

Reported-by: Namjae Jeon <linkinjeon@kernel.org>
Tested-by: Namjae Jeon <linkinjeon@kernel.org>
Cc: stable@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
2 months agoRevert "ALSA: hda: Add ASRock X670E Taichi to denylist"
Mario Limonciello (AMD) [Wed, 13 Aug 2025 14:03:08 +0000 (09:03 -0500)]
Revert "ALSA: hda: Add ASRock X670E Taichi to denylist"

On a motherboard with an AMD Granite Ridge CPU there is a report
that 3.5mm microphone and headphones aren't working.  In the
log it's observed:

snd_hda_intel 0000:02:00.6: Skipping the device on the denylist

This was because of commit df42ee7e22f03 ("ALSA: hda: Add ASRock
X670E Taichi to denylist").  Reverting this commit allows the
microphone and headphones to work again. As at least some combinations
of this motherboard do have applicable devices, revert so that they
can be probed.

Cc: Richard Gong <richard.gong@amd.com>
Cc: Juan Martinez <juan.martinez@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
Link: https://patch.msgid.link/20250813140427.1577172-1-superm1@kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2 months agoALSA: azt3328: Put __maybe_unused for inline functions for gameport
Takashi Iwai [Wed, 13 Aug 2025 15:36:27 +0000 (17:36 +0200)]
ALSA: azt3328: Put __maybe_unused for inline functions for gameport

Some inline functions are unused depending on kconfig, and the recent
change for clang builds made those handled as errors with W=1.
For avoiding pitfalls, mark those with __maybe_unused attributes.

Link: https://patch.msgid.link/20250813153628.12303-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2 months agoMerge tag 'mm-hotfixes-stable-2025-08-12-20-50' of git://git.kernel.org/pub/scm/linux...
Linus Torvalds [Wed, 13 Aug 2025 15:28:33 +0000 (08:28 -0700)]
Merge tag 'mm-hotfixes-stable-2025-08-12-20-50' of git://git./linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
 "12 hotfixes. 5 are cc:stable and the remainder address post-6.16
  issues or aren't considered necessary for -stable kernels.

  10 of these fixes are for MM"

* tag 'mm-hotfixes-stable-2025-08-12-20-50' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  proc: proc_maps_open allow proc_mem_open to return NULL
  mm/mremap: avoid expensive folio lookup on mremap folio pte batch
  userfaultfd: fix a crash in UFFDIO_MOVE when PMD is a migration entry
  mm: pass page directly instead of using folio_page
  selftests/proc: fix string literal warning in proc-maps-race.c
  fs/proc/task_mmu: hold PTL in pagemap_hugetlb_range and gather_hugetlb_stats
  mm/smaps: fix race between smaps_hugetlb_range and migration
  mm: fix the race between collapse and PT_RECLAIM under per-vma lock
  mm/kmemleak: avoid soft lockup in __kmemleak_do_cleanup()
  MAINTAINERS: add Masami as a reviewer of hung task detector
  mm/kmemleak: avoid deadlock by moving pr_warn() outside kmemleak_lock
  kasan/test: fix protection against compiler elision

2 months agoASoC: tas2781: Normalize the volume kcontrol name
Baojun Xu [Wed, 13 Aug 2025 10:07:08 +0000 (18:07 +0800)]
ASoC: tas2781: Normalize the volume kcontrol name

Change the name of the kcontrol from "Gain" to "Volume".

Signed-off-by: Baojun Xu <baojun.xu@ti.com>
Link: https://patch.msgid.link/20250813100708.12197-1-baojun.xu@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>