drm/xe/doc: Document device wedged and runtime survivability

Add documentation for vendor specific device wedged recovery method
and runtime survivability.

v2: fix documentation (Raag)
v3: add userspace tool for firmware update (Raag)
v4: use consistent documentation (Raag)
v5: add more documentation

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-8-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe/xe_survivability: Add support for Runtime survivability mode

Certain runtime firmware errors can cause the device to be in a unusable
state requiring a firmware flash to restore normal operation.
Runtime Survivability Mode indicates firmware flash is necessary by
wedging the device and exposing survivability mode sysfs.

The below sysfs is an indication that device is in survivability mode

/sys/bus/pci/devices/<device>/survivability_mode

v2: Fix kernel-doc (Umesh)
v3: Add user friendly dmesg (Frank)

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-7-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe/xe_survivability: Refactor survivability mode

Refactor survivability mode code to support both boot
and runtime survivability.

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-6-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe: Add a helper function to set recovery method

Add a helper function to set recovery method. The recovery
method can be set before declaring the device wedged and sending the
drm wedged uevent. If no method is set, default unbind/re-bind method
will be set.

v2: fix documentation (Raag)

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-5-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe: Set GT as wedged before sending wedged uevent

Userspace should be notified after setting the device as wedged.
Re-order function calls to set gt wedged before sending uevent.

Cc: Matthew Brost <matthew.brost@intel.com>
Suggested-by: Raag Jadav <raag.jadav@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-4-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm: Add a vendor-specific recovery method to drm device wedged uevent

Address the need for a recovery method (firmware flash on Firmware errors)
introduced in the later patches of Xe KMD.
Whenever XE KMD detects a firmware error, a firmware flash is required to
recover the device to normal operation.

The initial proposal to use 'firmware-flash' as a recovery method was
not applicable to other drivers and could cause multiple recovery
methods specific to vendors to be added.
To address this a more generic 'vendor-specific' method is introduced,
guiding users to refer to vendor specific documentation and system logs
for detailed vendor specific recovery procedure.

Add a recovery method 'WEDGED=vendor-specific' for such errors.
Vendors must provide additional recovery documentation if this method
is used.

It is the responsibility of the consumer to refer to the correct vendor
specific documentation and usecase before attempting a recovery.

For example: If driver is XE KMD, the consumer must refer
to the documentation of 'Device Wedging' under 'Documentation/gpu/xe/'.

v2: fix documentation (Raag)
v3: add more details to commit message (Sima, Rodrigo, Raag)
    add an example script to the documentation (Raag)
v4: use consistent naming (Raag)
v5: fix commit message
v6: add more documentation

Cc: André Almeida <andrealmeid@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona.vetter@ffwll.ch>
Cc: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250826063419.3022216-3-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe: Add documentation for Xe Device Wedging

Add documentation for Xe Device Wedging so that
file can be referenced in following patches.

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-2-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe/uapi: Add UAPI for querying VMA count and memory attributes

Introduce the DRM_IOCTL_XE_VM_QUERY_MEMORY_RANGE_ATTRS ioctl to allow
userspace to query memory attributes of VMAs within a user specified
virtual address range.

Userspace first calls the ioctl with num_mem_ranges = 0,
sizeof_mem_ranges_attr = 0 and vector_of_vma_mem_attr = NULL to retrieve
the number of memory ranges (vmas) and size of each memory range attribute.
Then, it allocates a buffer of that size and calls the ioctl again to fill
the buffer with memory range attributes.

This two-step interface allows userspace to first query the required
buffer size, then retrieve detailed attributes efficiently.

v2 (Matthew Brost)
- Use same ioctl to overload functionality

v3
- Add kernel-doc

v4
- Make uapi future proof by passing struct size (Matthew Brost)
- make lock interruptible (Matthew Brost)
- set reserved bits to zero (Matthew Brost)
- s/__copy_to_user/copy_to_user (Matthew Brost)
- Avod using VMA term in uapi (Thomas)
- xe_vm_put(vm) is missing (Shuicheng)

v5
- Nits
- Fix kernel-doc

Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Shuicheng Lin <shuicheng.lin@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-21-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe: Enable madvise ioctl for xe

Ioctl enables setting up of memory attributes in user provided range.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-20-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe: Reset VMA attributes to default in SVM garbage collector

Restore default memory attributes for VMAs during garbage collection
if they were modified by madvise. Reuse existing VMA if fully overlapping;
otherwise, allocate a new mirror VMA.

v2 (Matthew Brost)
- Add helper for vma split
- Add retry to get updated vma

v3
- Rebase on gpuvm layer

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-19-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/vm: Add helper to check for default VMA memory attributes

Introduce a new helper function `xe_vma_has_default_mem_attrs()` to
determine whether a VMA's memory attributes are set to their default
values. This includes checks for atomic access, PAT index, and preferred
location.

Also, add a new field `default_pat_index` to `struct xe_vma_mem_attr`
to track the initial PAT index set during the first bind. This helps
distinguish between default and user-modified pat index, such as those
changed via madvise.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-18-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/madvise: Skip vma invalidation if mem attr are unchanged

If a VMA within the madvise input range already has the same memory
attribute as the one requested by the user, skip PTE zapping for that
VMA to avoid unnecessary invalidation.

v2 (Matthew Brost)
- fix skip_invalidation for new attributes
- s/u32/bool
- Remove unnecessary assignment  for kzalloc'ed

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-17-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/bo: Update atomic_access attribute on madvise

Update the bo_atomic_access based on user-provided input and determine
the migration to smem during a CPU fault

v2 (Matthew Brost)
- Avoid cpu unmapping if bo is already in smem
- check atomics on smem too for ioctl
- Add comments

v3
- Avoid migration in prefetch

v4 (Matthew Brost)
- make sanity check function bool
- add assert for smem placement
- fix doc

v5 (Matthew Brost)
- NACK atomic fault with  DRM_XE_ATOMIC_CPU

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-16-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/bo: Add attributes field to xe_bo

A single BO can be linked to multiple VMAs, making VMA attributes
insufficient for determining the placement and PTE update attributes
of the BO. To address this, an attributes field has been added to the
BO.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-15-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/svm: Consult madvise preferred location in prefetch

When prefetch region is DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC, prefetch svm
ranges to preferred location provided by madvise.

v2 (Matthew Brost)
- Fix region, devmem_fd usages
- consult madvise is applicable for other vma's too.

v3
- Fix atomic handling

v4
- Fix xe_svm_range_validate to check for
  DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC too.

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-14-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/uapi: Add flag for consulting madvise hints on svm prefetch

Introduce flag DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC to ensure prefetching
in madvise-advised memory regions

v2 (Matthew Brost)
- Add kernel-doc

v3 (Matthew Brost)
- Fix kernel-doc

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-13-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/svm: Support DRM_XE_SVM_MEM_RANGE_ATTR_PAT memory attribute

This attributes sets the pat_index for the svm used vma range, which is
utilized to ascertain the coherence.

v2 (Matthew Brost)
- Pat index sanity check

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-12-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/madvise: Update migration policy based on preferred location

When the user sets the valid devmem_fd as a preferred location, GPU fault
will trigger migration to tile of device associated with devmem_fd.

If the user sets an invalid devmem_fd the preferred location is current
placement(smem) only.

v2(Matthew Brost)
- Default should be faulting tile
- remove devmem_fd used as region

v3 (Matthew Brost)
- Add migration_policy
- Fix return condition
- fix migrate condition

v4
-Rebase

v5
- Add check for userptr and bo based vmas

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-11-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/svm: Add svm ranges migration policy on atomic access

If the platform does not support atomic access on system memory, and the
ranges are in system memory, but the user requires atomic accesses on
the VMA, then migrate the ranges to VRAM. Apply this policy for prefetch
operations as well.

v2
- Drop unnecessary vm_dbg

v3 (Matthew Brost)
- fix atomic policy
- prefetch shouldn't have any impact of atomic
- bo can be accessed from vma, avoid duplicate parameter

v4 (Matthew Brost)
- Remove TODO comment
- Fix comment
- Dont allow gpu atomic ops when user is setting atomic attr as CPU

v5 (Matthew Brost)
- Fix atomic checks
- Add userptr checks

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-10-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe: Implement madvise ioctl for xe

This driver-specific ioctl enables UMDs to control the memory attributes
for GPU VMAs within a specified input range. If the start or end
addresses fall within an existing VMA, the VMA is split accordingly. The
attributes of the VMA are modified as provided by the users. The old
mappings of the VMAs are invalidated, and TLB invalidation is performed
if necessary.

v2(Matthew brost)
- xe_vm_in_fault_mode can't be enabled by Mesa, hence allow ioctl in non
fault mode too
- fix tlb invalidation skip for same ranges in multiple op
- use helper for tlb invalidation
- use xe_svm_notifier_lock/unlock helper
- s/lockdep_assert_held/lockdep_assert_held_write
- Add kernel-doc

v3(Matthew Brost)
- make vfunc fail safe
- Add sanitizing input args before vfunc

v4(Matthew Brost/Shuicheng)
- Make locks interruptable
- Error handling fixes
- vm_put fixes

v5(Matthew Brost)
- Flush garbage collector before any locking.
- Add check for null vma

Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-9-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/svm: Add xe_svm_ranges_zap_ptes_in_range() for PTE zapping

Introduce xe_svm_ranges_zap_ptes_in_range(), a function to zap page table
entries (PTEs) for all SVM ranges within a user-specified address range.

-v2 (Matthew Brost)
Lock should be called even for tlb_invalidation

v3(Matthew Brost)
- Update comment
- s/notifier->itree.start/drm_gpusvm_notifier_start
- s/notifier->itree.last + 1/drm_gpusvm_notifier_end
- use WRITE_ONCE

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-8-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe: Allow CPU address mirror VMA unbind with gpu bindings for madvise

In the case of the MADVISE ioctl, if the start or end addresses fall
within a VMA and existing SVM ranges are present, remove the existing
SVM mappings. Then, continue with ops_parse to create new VMAs by REMAP
unmapping of old one.

v2 (Matthew Brost)
- Use vops flag to call unmapping of ranges in vm_bind_ioctl_ops_parse
- Rename the function

v3
- Fix doc

v4
- check if range is already in garbage collector (Matthew Brost)

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-7-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/svm: Split system allocator vma incase of madvise call

If the start or end of input address range lies within system allocator
vma split the vma to create new vma's as per input range.

v2 (Matthew Brost)
- Add lockdep_assert_write for vm->lock
- Remove unnecessary page aligned checks
- Add kerrnel-doc and comments
- Remove unnecessary unwind_ops and return

v3
- Fix copying of attributes

v4
- Nit fixes

v5
- Squash identifier for madvise in xe_vma_ops to this patch

v6/v7/v8
- Rebase on drm_gpuvm changes

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-6-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/vma: Modify new_vma to accept struct xe_vma_mem_attr as parameter

This change simplifies the logic by ensuring that remapped previous or
next VMAs are created with the same memory attributes as the original VMA.
By passing struct xe_vma_mem_attr as a parameter, we maintain consistency
in memory attributes.

-v2
 *dst = *src (Matthew Brost)

-v3 (Matthew Brost)
 Drop unnecessary helper
 pass attr ptr as input to new_vma and vma_create

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-5-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/vma: Move pat_index to vma attributes

The PAT index determines how PTEs are encoded and can be modified by
madvise. Therefore, it is now part of the vma attributes.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-4-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/vm: Add attributes struct as member of vma

The attribute of xe_vma will determine the migration policy and the
encoding of the page table entries (PTEs) for that vma.
This attribute helps manage how memory pages are moved and how their
addresses are translated. It will be used by madvise to set the
behavior of the vma.

v2 (Matthew Brost)
- Add docs

v3 (Matthew Brost)
- Add uapi references
- 80 characters line wrap

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-3-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

drm/xe/uapi: Add madvise interface

This commit introduces a new madvise interface to support
driver-specific ioctl operations. The madvise interface allows for more
efficient memory management by providing hints to the driver about the
expected memory usage and pte update policy for gpuvma.

v2 (Matthew/Thomas)
- Drop num_ops support
- Drop purgeable support
- Add kernel-docs
- IOWR/IOW

v3 (Matthew/Thomas)
- Reorder attributes
- use __u16 for migration_policy
- use __u64 for reserved in unions
- Avoid usage of vma

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-2-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>

Merge drm/drm-next into drm-xe-next

Sync with drm-misc-next which is necessary for changes in gpuvm
and gpusvm that will be used in xe.

Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe: switch to local xbasename() helper

Commit b0a2ee5567ab ("drm/xe: prepare xe_gen_wa_oob to be multi-use")
introduced a call to basename(). The GNU version of this function is not
portable and fails to build with alternative libc implementations like
musl or bionic. This causes the following build error:

  drivers/gpu/drm/xe/xe_gen_wa_oob.c:130:12: error: assignment to ‘const char *’ from ‘int’ makes pointer from integer without a cast [-Wint-conversion]
    130 |         fn = basename(fn);
        |            ^

While a POSIX version of basename() could be used, it would require a
separate header plus the behavior differs from GNU version in that it
might modify its argument. Not great.

Instead, implement a local xbasename() helper based on strrchr() that
provides the same functionality and avoids portability issues.

Fixes: b0a2ee5567ab ("drm/xe: prepare xe_gen_wa_oob to be multi-use")
Suggested-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Tiffany Yang <ynaffit@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250825155743.1132433-1-cmllamas@google.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe: Don't trigger rebind on initial dma-buf validation

On the first validate of an imported dma-buf (initial bind), the device
has no GPU mappings, so a rebind is unnecessary. Rebinding here is
harmful in multi-GPU setups and for VMs using preempt-fence mode, as it
would evict in-flight GPU work.

v2:
 - Drop dma_buf_validated, check for XE_PL_SYSTEM (Thomas)

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250825152841.3837378-1-matthew.brost@intel.com

drm/xe/vm: Clear the scratch_pt pointer on error

Avoid triggering a dereference of an error pointer on cleanup in
xe_vm_free_scratch() by clearing any scratch_pt error pointer.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Fixes: 06951c2ee72d ("drm/xe: Use NULL PTEs as scratch PTEs")
Cc: Brian Welty <brian.welty@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: <stable@vger.kernel.org> # v6.8+
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821143045.106005-4-thomas.hellstrom@linux.intel.com

drm/xe/tests/xe_dma_buf: Set the drm_object::dma_buf member

This member is set when exporting using prime. However
the xe_gem_prime_export() alone doesn't set it, since it's done
later in the prime export flow.
For the test, set it manually and remove the hack that set it
temporarily when it was really needed.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821143045.106005-3-thomas.hellstrom@linux.intel.com

drm/xe/vm: Don't pin the vm_resv during validation

The pinning has the odd side-effect that unlocking *any* resv
during validation triggers an "unlocking pinned lock" warning.

Cc: Matthew Brost <matthew.brost@intel.com>
Fixes: 9d5558649f68 ("drm/xe: Rework eviction rejection of bound external bos")
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821143045.106005-2-thomas.hellstrom@linux.intel.com

drm/xe/xe_sync: avoid race during ufence signaling

Marking ufence as signalled after copy_to_user() is too late.
Worker thread which signals ufence by memory write might be raced
with another userspace vm-bind call. In map/unmap scenario unmap
may still see ufence is not signalled causing -EBUSY. Change the
order of marking / write to user-fence fixes this issue.

Fixes: 977e5b82e090 ("drm/xe: Expose user fence from xe_sync_entry")
Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5536
Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250820083903.2109891-2-zbigniew.kempczynski@intel.com

Merge tag 'drm-misc-next-2025-08-21' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next

drm-misc-next for v6.18:

Core Changes:

bridge:
- Support Content Protection property

gpuvm:
- Support madvice in Xe driver

mipi:
- Add more multi-read/write helpers for improved error handling

Driver Changes:

amdxdna:
- Refactoring wrt. hardware contexts

bridge:
- display-connector: Improve DP display detection

panel:
- Fix includes in various drivers

panthor:
- Add support for Mali G710, G510, G310, Gx15, Gx20, Gx25
- Improve cache flushing

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250821073822.GA45904@2a02-2454-fd5e-fd00-8f09-b5f-980b-a7ef.dyn6.pyur.net

drm/xe/configfs: Dump custom settings when binding

Device configuration using configfs could be prepared long time prior
the driver load. Currently all the xe configfs entries are for things
that are important to have in the log if a non-default value is being
used. Add a info-level message about that with the individual entries
that are different than the default.

Based on previous patch by Michal Wajdeczko.

Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-12-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/configfs: Minor fixes to documentation

Add a few missing punctuation and line breaks and make the syntax for
code snippets common to all of them.

Reviewed-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-11-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/configfs: Improve documentation steps

The steps are roughly:

1. Load the module without binding to the device
2. Configure the desired device
3. Bind the device

Move the binding part to the "Create devices" since it's not exclusive
to the survivability_mode attribute and better document the steps.

Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-10-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/configfs: Use tree-like output in documentation

When documenting the directories, use an output similar to the `tree`
command and add VFs and missing attributes.

Reviewed-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-9-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/configfs: Use guard() for dev->lock

Instead of the manual lock()/unlock() pattern, use guard() which will
make things easier for handling errors or early returns.

Reviewed-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-8-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/configfs: Allow to enable PSMI

Now that additional WAs are in place and it's possible to allocate
buffers through debugfs, add the configfs attribute to turn PSMI on.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-7-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/configfs: Simplify kernel doc

From the caller perspective reading the documentation, there's no need
to be so specific about everything the function is doing/checking. Just
document the functionality a caller cares about.

Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-6-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/psmi: Add Wa_16023683509

This WA ensures GuC will restore the media MCFG registers at C6
exit.

Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-5-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/psmi: Add Wa_14020001231

Enable Wa 14020001231 to block psmi interrupts during C6 entry exit
flow. It's only enabled if PSMI is enabled in runtime.

Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-4-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/rtp: Add match for psmi

Add match to be used on WAs for only enabling workarounds if psmi is
intended to be used.

Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-3-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/psmi: Add debugfs interface for PSMI

Requirement for PSMI capture is to have a physically contiguous buffer.
All the needed configuration is done by the userspace tool directly to
the GPU via mmio access.

This interface only support allocating from VRAM regions. For integrated
devices, the PSMI buffer is in SYSTEM memory and should be allocated by
userspace using hugetlbfs.

Here we add the ability to allocate a region of physically contiguous
memory by writing to debugfs file (listed below). For multi-tile devices,
the capture tool requires ability to allocate a capture buffer per tile
(VRAM region) and so user can specify a region_mask. The tool then
can mmap the buffers via direct mmap of the PCIBAR via sysfs.

To support the capture tool, 3 new debugfs entries are added:

   psmi_capture_addr - physical address per VRAM region's capture buffer
   psmi_capture_region_mask - select which region(s) to allocate a buffer
   psmi_capture_size - size of current capture buffer

Writing psmi_capture_size will allocate new buffer of requested size per
region after freeing any current buffers.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Original-author: Brian Welty <brian.welty@intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com> # v2
Link: https://lore.kernel.org/r/20250821-psmi-v5-2-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/psmi: Add GuC flag to enable PSMI

PSMI allows to capture data from the GPU useful for early
validation. From the kernel side there isn't much to be done, just a few
things:

1) Toggle the feature support in GuC
2) Enable some additional WAs
3) Allocate buffers

Here is the first step, with the next ones to follow. For now everything
is disabled through a check in configfs that is currently hardcoded to
disabled.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-1-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/xe/pcode: Initialize data0 for pcode read routine

There are two registers filled in when reading data from
pcode besides the mailbox itself. Currently, we allow a NULL
value for the second of these two (data1) and assume the first
is defined. However, many of the routines that are calling
this function assume that pcode will ignore the value being
passed in and so leave that first value (data0) defined but
uninitialized. To be safe, make sure this value is always
initialized to something (0 generally) in the event pcode
behavior changes and starts using this value.

v2: Fix sob/author

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250819201054.393220-1-stuart.summers@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe/kunit: Extend platform generator with PTL

Our list of typical platforms used to generate test device objects
does not contain any PANTHERLAKE. Add one.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250818192032.633-1-michal.wajdeczko@intel.com

drm/xe: Use for_each_gt to define gt_count

We are currently bumping gt_count as we define GTs for each tile. While
that works with our current code, there are reasons to improve that:

* That section of the code is dedicated to define each tile's set of GTs
  and their respective info. The gt_count can be seen as a device level
  property. While it is fair to bump it as we define each GT, we can
  also focus on the GT themselves and count after we are done with the
  definitions.

* More *importantly*, gt_count should reflect the number of GTs that we
  would get when looping over them with for_each_gt(). Bumping gt_count
  the way we are currently doing makes that value not really connected
  to for_each_gt(). This could bite us in the future if in the loop gets
  extra checks that are not accounted for in each existing "gt_count++".

As such, let's use for_each_gt() and extract the calculation of gt_count
into a separate block, just after we define the set of GTs for each
tile.

Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250818-gt_count-improvements-v4-2-ee12870c6f57@intel.com
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>

drm/xe: Probe for tile count during device info initialization

The function mmio_multi_tile_setup() does not look like the proper
location for probing for the number of existing tiles in the hardware.
It should not be that function's responsibility and such information
should instead be already available when it gets called.

The fact that we need to adjust gt_count is a symptom of that.

Move probing of available tile count to a dedicated function named
xe_info_probe_tile_count() and call it from xe_info_init(), which seems
like a more appropriate place for that.

With that move, we no longer need to (and shouldn't) adjust gt_count as
a part of xe_info_probe_tile_count(), as that field will be initialized
later in xe_info_init().

v4:
  - Only probe for tile count if the default tile_count != 1 (just like
    was done in mmio_multi_tile_setup()). (CI)
v3:
  - Unchanged.
v2:
  - Use KUnit static stub so that we do not try to query hardware when
    running KUnit tests. (CI)
  - Tweak last paragraph of commit message to make it clearer.
    (Jonathan)

Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250818-gt_count-improvements-v4-1-ee12870c6f57@intel.com
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>

drm/xe/tuning: Apply "Disable NULL query for Anyhit Shader" to Xe2

Extend the "Disable NULL query for Anyhit Shader" tuning to Xe2
(graphics version 2000+) platforms, in addition to Xe3.

This sets the DIS_NULL_QUERY bit in RT_CTRL to disable null query
for Anyhit shaders on both Xe2 and Xe3.

This is a change in behavior that can regress a userspace not
prepared for it. However it's not feasible to change dynamically
the option per client or per exec queue via an opt-in flag. Mesa
is already prepared for that and it got propagated to their
stable versions. Even if it was possible, at this point adding a
flag would mean mesa would also need to propagate such a flag to
their stable versions, otherwise the previous fix would not be
used.

Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Link: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35044
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35044
Link: https://lore.kernel.org/r/20250819061151.1272622-1-nitin.r.gote@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

drm/bridge: anx7625: register content protect property

Set the `support_hdcp` bit to enable the connector to register content
protection during initialization.

Signed-off-by: Hsin-Yi Wang <hsinyi@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Fei Shao <fshao@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250812082135.3351172-3-fshao@chromium.org

drm_bridge: register content protect property

Some bridges can update HDCP status based on userspace requests if they
support HDCP.

The HDCP property is created after connector initialization and before
registration, just like other connector properties.

Add the content protection property to the connector if a bridge
supports HDCP.

Signed-off-by: Hsin-Yi Wang <hsinyi@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Fei Shao <fshao@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250812082135.3351172-2-fshao@chromium.org

drm/panel: panel-samsung-s6e88a0-ams427ap24: Fix includes

Include <linux/property.h> to declare device_property_read_bool() and
<linux/mod_devicetable.h> to declare struct of_device_id. Avoids the
dependency on the backlight header to include both.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250812082509.227879-1-tzimmermann@suse.de

drm/virtio: clean up minor codestyle issues

Fix codestyle warnings and errors generated by CHECKPATCH in virtio
source files.

Signed-off-by: Athul Raj Kollareth <krathul3152@gmail.com>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Link: https://lore.kernel.org/r/20250813062109.5326-1-krathul3152@gmail.com

Merge drm/drm-next into drm-misc-next

Bring v6.17-rc2 in to unstuck for-linux-next.

Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/gpusvm: Make drm_gpusvm_for_each_* macros public

The drm_gpusvm_for_each_notifier, drm_gpusvm_for_each_notifier_safe and
drm_gpusvm_for_each_range_safe macros are useful for locating notifiers
and ranges within a user-specified range. By making these macros public,
we enable broader access and utility for developers who need to leverage
them in their implementations.

v2 (Matthew Brost)
- drop inline __drm_gpusvm_range_find
- /s/notifier_iter_first/drm_gpusvm_notifier_find

Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-5-himal.prasad.ghimiray@intel.com

drm/gpuvm: Introduce drm_gpuvm_madvise_ops_create

This ops is used to iterate over GPUVA's in the user-provided range
and split the existing sparse VMA's if the start or end of the input
range lies within it. The operations can create up to 2 REMAPS and 2 MAPs.

The primary use case is for drivers to assign attributes to GPU VAs in
the specified range without performing unmaps or merging mappings,
supporting fine-grained control over sparse va's.

Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Boris Brezillon <bbrezillon@kernel.org>
Cc: <dri-devel@lists.freedesktop.org>
Signed-off-by: Himal Prasad Ghimiray<himal.prasad.ghimiray@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-4-himal.prasad.ghimiray@intel.com

drm/gpuvm: Kill drm_gpuva_init()

drm_gpuva_init() only has one internal user, and given we are about to
add new optional fields, it only add maintenance burden for no real
benefit, so let's kill the thing now.

Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Rob Clark <robin.clark@oss.qualcomm.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Rob Clark <robin.clark@oss.qualcomm.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-3-himal.prasad.ghimiray@intel.com

drm/gpuvm: Pass map arguments through a struct

We are about to pass more arguments to drm_gpuvm_sm_map[_ops_create](),
so, before we do that, let's pass arguments through a struct instead
of changing each call site every time a new optional argument is added.

Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Brendan King <Brendan.King@imgtec.com>
Cc: Matt Coster <matt.coster@imgtec.com>
Cc: Boris Brezillon <bbrezillon@kernel.org>
Cc: Caterina Shablia <caterina.shablia@collabora.com>
Cc: Rob Clark <robin.clark@oss.qualcomm.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: <dri-devel@lists.freedesktop.org>
Co-developed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Matt Coster <matt.coster@imgtec.com> # imagination/pvr_vm.c
Acked-by: Matt Coster <matt.coster@imgtec.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-2-himal.prasad.ghimiray@intel.com

rust: drm: update ARef and AlwaysRefCounted imports from sync::aref

Update call sites in drm to import `ARef` and
`AlwaysRefCounted` from `sync::aref` instead of `types`.

This aligns with the ongoing effort to move `ARef` and
`AlwaysRefCounted` to sync.

Suggested-by: Benno Lossin <lossin@kernel.org>
Link: https://github.com/Rust-for-Linux/linux/issues/1173
Signed-off-by: Shankari Anand <shankari.ak0208@gmail.com>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Reviewed-by: Elle Rhumsaa <elle@weathered-steel.dev>
Link: https://lore.kernel.org/r/20250815161706.1324860-1-shankari.ak0208@gmail.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>

Merge tag 'drm-misc-next-2025-08-14' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next

drm-misc-next for v6.18:

UAPI Changes:

- Add DRM_IOCTL_GEM_CHANGE_HANDLE for reassigning GEM handles
- Document DRM_MODE_PAGE_FLIP_EVENT

Cross-subsystem Changes:

fbcon:
- Add missing declarations in fbcon.h

Core Changes:

bridge:
- Fix ref counting

panel:
- Replace and remove mipi_dsi_generic_write_{seq/_chatty}()

sched:
- Fixes

Rust:
- Drop Opaque<> from ioctl arguments

Driver Changes:

amdxdma:
- Support buffers allocated by user space
- Streamline PM interfaces
- Fixes

bridge:
- cdns-dsi: Various improvements to mode setting
- Support Solomon SSD2825 plus DT bindings
- Support Waveshare DSI2DPI plus DT bindings

gud:
- Fixes

ivpu:
- Fixes

nouveau:
- Use GSP firmware by default
- Fixes

panel:
- panel-edp: Support mt8189 Chromebooks; Support BOE NV140WUM-N64;
  Support SHP LQ134Z1; Fixes
- panel-simple: Support Olimex LCD-OLinuXino-5CTS plus DT bindings
- Support Samsung AMS561RA01
- Support Hydis HV101HD1 plus DT bindings

panthor:
- Print task/pid on errors
- Fixes

renesas:
- convert to RUNTIME_PM_OPS

repaper:
- Use shadow-plane helpers

rocket:
- Add driver for Rockchip NPU plus DT bindings

sharp-memory:
- Use shadow-plane helpers

simpledrm:
- Use of_reserved_mem_region_to_resource() helper

tidss:
- Use crtc_ fields for programming display mode
- Remove other drivers from aperture

v3d:
- Support querying nubmer of GPU resets for KHR_robustness

vmwgfx:
- Fixes

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250814072454.GA18104@linux.fritz.box

drm/xe: Make page size consistent in loop

If PAGE_SIZE != XE_PAGE_SIZE (which is currently locked behind
CONFIG_BROKEN), this would generate the wrong number of PDEs.

Since these PDEs are consumed by the GPU, the GPU page size needs to be
used.

Signed-off-by: Simon Richter <Simon.Richter@hogyros.de>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250818064806.2835-1-Simon.Richter@hogyros.de

drm/xe: Untangle vm_bind_ioctl cleanup order

Change the resource cleanup order to align with reverse allocation order.

Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Christoph Manszewski <christoph.manszewski@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250813101231.196632-3-christoph.manszewski@intel.com

drm/xe: Fix vm_bind_ioctl double free bug

If the argument check during an array bind fails, the bind_ops are freed
twice as seen below. Fix this by setting bind_ops to NULL after freeing.

==================================================================
BUG: KASAN: double-free in xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
Free of addr ffff88813bb9b800 by task xe_vm/14198

CPU: 5 UID: 0 PID: 14198 Comm: xe_vm Not tainted 6.16.0-xe-eudebug-cmanszew+ #520 PREEMPT(full)
Hardware name: Intel Corporation Alder Lake Client Platform/AlderLake-P DDR5 RVP, BIOS ADLPFWI1.R00.2411.A02.2110081023 10/08/2021
Call Trace:
 <TASK>
 dump_stack_lvl+0x82/0xd0
 print_report+0xcb/0x610
 ? __virt_addr_valid+0x19a/0x300
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 kasan_report_invalid_free+0xc8/0xf0
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 check_slab_allocation+0x102/0x130
 kfree+0x10d/0x440
 ? should_fail_ex+0x57/0x2f0
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe]
 ? __lock_acquire+0xab9/0x27f0
 ? lock_acquire+0x165/0x300
 ? drm_dev_enter+0x53/0xe0 [drm]
 ? find_held_lock+0x2b/0x80
 ? drm_dev_exit+0x30/0x50 [drm]
 ? drm_ioctl_kernel+0x128/0x1c0 [drm]
 drm_ioctl_kernel+0x128/0x1c0 [drm]
 ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe]
 ? find_held_lock+0x2b/0x80
 ? __pfx_drm_ioctl_kernel+0x10/0x10 [drm]
 ? should_fail_ex+0x57/0x2f0
 ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe]
 drm_ioctl+0x352/0x620 [drm]
 ? __pfx_drm_ioctl+0x10/0x10 [drm]
 ? __pfx_rpm_resume+0x10/0x10
 ? do_raw_spin_lock+0x11a/0x1b0
 ? find_held_lock+0x2b/0x80
 ? __pm_runtime_resume+0x61/0xc0
 ? rcu_is_watching+0x20/0x50
 ? trace_irq_enable.constprop.0+0xac/0xe0
 xe_drm_ioctl+0x91/0xc0 [xe]
 __x64_sys_ioctl+0xb2/0x100
 ? rcu_is_watching+0x20/0x50
 do_syscall_64+0x68/0x2e0
 entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7fa9acb24ded

Fixes: b43e864af0d4 ("drm/xe/uapi: Add DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Christoph Manszewski <christoph.manszewski@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250813101231.196632-2-christoph.manszewski@intel.com

accel/amdxdna: Add a function to walk hardware contexts

Walking hardware contexts created by a process is duplicated in multiple
spots. Add a function, amdxdna_hwctx_walk(), and replace all spots.

hwctx_srcu and dev_lock are good enough to protect hardware context list.
Remove hwctx_lock.

Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250815171634.3417487-1-lizhi.hou@amd.com

Linux 6.17-rc2

drm/xe/pm: Disable RPM for SR-IOV VFs

VFs without native PCIe Power Management (PM) capabilities inherit their
PF's power state as per PCIe specifications(§5.10.1 PCIe Base Spec 7.0).
Enabling Runtime Power Management (RPM) for these VFs trigger unnecessary
driver suspend/resume operations that ultimately perform no PCI-level power
transition.

Since VFs without PM capabilities cannot independently enter low-power
states, the existing RPM workflow becomes redundant:
1. Driver executes full suspend/resume sequence
2. PCI PM transition step becomes no-op
3. VF power state remains tied to PF's status

Disabling RPM for VFs eliminates this redundant processing while
maintaining proper power management through PF dependency. This
optimization ensures VFs follow their PF's power state without superfluous
runtime handling.

Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Anshuman Gupta <anshuman.gupta@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250812163613.9954-1-satyanarayana.k.v.p@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drm/xe/i2c: Introduce xe_i2c_present()

In preparation of wider usecases which require checking for I2C
controller presence, introduce xe_i2c_present() helper.

v2: Doc fix (Riana)

Suggested-by: Riana Tauro <riana.tauro@intel.com>
Signed-off-by: Raag Jadav <raag.jadav@intel.com>
Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250814091845.1356135-1-raag.jadav@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

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

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()

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

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

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

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

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

drm/bridge: display-connector: don't set OP_DETECT for DisplayPorts

Detecting the monitor for DisplayPort targets is more complicated than
just reading the HPD pin level: it requires reading the DPCD in order to
check what kind of device is attached to the port and whether there is
an actual display attached.

In order to let DRM framework handle such configurations, disable
DRM_BRIDGE_OP_DETECT for dp-connector devices, letting the actual DP
driver perform detection. This still keeps DRM_BRIDGE_OP_HPD enabled, so
it is valid for the bridge to report HPD events.

Currently inside the kernel there are only two targets which list
hpd-gpios for dp-connector devices: arm64/qcom/qcs6490-rb3gen2 and
arm64/qcom/sa8295p-adp. Both should be fine with this change.

Cc: Bjorn Andersson <andersson@kernel.org>
Cc: Konrad Dybcio <konradybcio@kernel.org>
Cc: linux-arm-msm@vger.kernel.org
Acked-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Link: https://lore.kernel.org/r/20250802-dp-conn-no-detect-v1-1-2748c2b946da@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>

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

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

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

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

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

drm/panfrost: Remove redundant pm_runtime_mark_last_busy() calls

pm_runtime_put_autosuspend(), pm_runtime_put_sync_autosuspend(),
pm_runtime_autosuspend() and pm_request_autosuspend() now include a call
to pm_runtime_mark_last_busy(). Remove the now-reduntant explicit call to
pm_runtime_mark_last_busy().

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250704075411.3218059-1-sakari.ailus@linux.intel.com

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

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"

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

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)

drm/panthor: Add support for Mali-Gx20 and Mali-Gx25 GPUs

This patch adds firmware binary and GPU model naming support for
Mali-Gx20 and Mali-Gx25 GPUs.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-8-karunika.choo@arm.com

drm/panthor: Make MMU cache maintenance use FLUSH_CACHES command

As the FLUSH_MEM and FLUSH_PT MMU_AS commands are deprecated in GPUs
from Mali-Gx20 onwards, this patch adds support for performing cache
maintenance via the FLUSH_CACHES command in GPU_COMMAND in place of
FLUSH_MEM and FLUSH_PT commands.

Mali-Gx10 and Mali-Gx15 GPUs also has support for the FLUSH_CACHES
command and will also use this by default going forward.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-7-karunika.choo@arm.com

drm/panthor: Serialize GPU cache flush operations

In certain scenarios, it is possible for multiple cache flushes to be
requested before the previous one completes. This patch introduces the
cache_flush_lock mutex to serialize these operations and ensure that
any requested cache flushes are completed instead of dropped.

Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Co-developed-by: Dennis Tsiang <dennis.tsiang@arm.com>
Signed-off-by: Dennis Tsiang <dennis.tsiang@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-6-karunika.choo@arm.com

drm/panthor: Add support for Mali-Gx15 family of GPUs

Mali-Gx15 introduces a new GPU_FEATURES register that provides
information about GPU-wide supported features. The register value will
be passed on to userspace via gpu_info.

Additionally, Mali-Gx15 presents an 'Immortalis' naming variant
depending on the shader core count and presence of Ray Intersection
feature support.

This patch adds:
- support for correctly identifying the model names for Mali-Gx15 GPUs.
- arch 11.8 FW binary support

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-5-karunika.choo@arm.com

drm/panthor: Add support for Mali-G710, Mali-G510 and Mali-G310

This patch adds GPU model name and FW binary support for Mali-G710,
Mali-G510, and Mali-G310.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-4-karunika.choo@arm.com

drm/panthor: Simplify getting the GPU model name

This patch replaces the panthor_model structure with a simple switch
case based on the product_id which is in the format of:
        ((arch_major << 24) | product_major)

This simplifies comparison and allows extending of the function to
accommodate naming differences based on supported GPU features.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-3-karunika.choo@arm.com

drm/panthor: Add panthor_hw and move gpu_info initialization into it

This patch introduces panthor_hw and moves the initialization of the
gpu_info struct into panthor_hw.c in preparation of handling future GPU
register and naming changes.

Future GPU support can be added by extending panthor_gpu_info_init()
with the necessary register reads behind GPU architecture version guards
if the change is minor. For more complex changes, the function can be
forked and the appropriate function will need to be called based on the
GPU architecture version.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-2-karunika.choo@arm.com

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

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

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

drm/panel: novatek-nt35560: Clean up driver

Update driver to use the "multi" variants of MIPI functions which
facilitate improved error handling and cleaner driver code.

Remove information from a comment which was made obsolete by commit
994ea402c767 ("drm/panel: Rename Sony ACX424 to Novatek NT35560"), which
determined that this driver supports the Novatek NT35560 panel
controller.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731032343.1258366-4-me@brighamcampbell.com

drm: Add MIPI read_multi func and two write macros

Create mipi_dsi_dcs_read_multi(), which accepts a mipi_dsi_multi_context
struct for improved error handling and cleaner panel driver code.

Create mipi_dsi_dcs_write_var_seq_multi() and
mipi_dsi_generic_write_var_seq_multi() macros which allow MIPI panel
drivers to write non-constant data to display controllers.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731032343.1258366-3-me@brighamcampbell.com