linux-stable.git/drivers/dax, branch master

Merge tag 'libnvdimm-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

2026-04-21T21:12:01+00:00

Pull dax updates from Ira Weiny:
 "The series adds DAX support required for the upcoming fuse/famfs file
  system.[1] The support here is required because famfs is backed by
  devdax rather than pmem. This all lays the groundwork for using shared
  memory as a file system"

Link: https://lore.kernel.org/all/0100019d43e5f632-f5862a3e-361c-4b54-a9a6-96c242a8f17a-000000@email.amazonses.com/ [1]

* tag 'libnvdimm-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  dax/fsdev: fix uninitialized kaddr in fsdev_dax_zero_page_range()
  dax: export dax_dev_get()
  dax: Add fs_dax_get() func to prepare dax for fs-dax usage
  dax: Add dax_set_ops() for setting dax_operations at bind time
  dax: Add dax_operations for use by fs-dax on fsdev dax
  dax: Save the kva from memremap
  dax: add fsdev.c driver for fs-dax on character dax
  dax: Factor out dax_folio_reset_order() helper
  dax: move dax_pgoff_to_phys from [drivers/dax/] device.c to bus.c

Merge tag 'cxl-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl

2026-04-17T22:52:58+00:00

Pull CXL (Compute Express Link) updates from Dave Jiang:
 "The significant change of interest is the handling of soft reserved
  memory conflict between CXL and HMEM. In essence CXL will be the first
  to claim the soft reserved memory ranges that belongs to CXL and
  attempt to enumerate them with best effort. If CXL is not able to
  enumerate the ranges it will punt them to HMEM.

  There are also MAINTAINERS email changes from Dan Williams and
  Jonathan Cameron"

* tag 'cxl-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl: (37 commits)
  MAINTAINERS: Update Jonathan Cameron's email address
  cxl/hdm: Add support for 32 switch decoders
  MAINTAINERS: Update address for Dan Williams
  tools/testing/cxl: Enable replay of user regions as auto regions
  cxl/region: Add a region sysfs interface for region lock status
  tools/testing/cxl: Test dax_hmem takeover of CXL regions
  tools/testing/cxl: Simulate auto-assembly failure
  dax/hmem: Parent dax_hmem devices
  dax/hmem: Fix singleton confusion between dax_hmem_work and hmem devices
  dax/hmem: Reduce visibility of dax_cxl coordination symbols
  cxl/region: Constify cxl_region_resource_contains()
  cxl/region: Limit visibility of cxl_region_contains_resource()
  dax/cxl: Fix HMEM dependencies
  cxl/region: Fix use-after-free from auto assembly failure
  cxl/core: Check existence of cxl_memdev_state in poison test
  cxl/core: use cleanup.h for devm_cxl_add_dax_region
  cxl/core/region: move dax region device logic into region_dax.c
  cxl/core/region: move pmem region driver logic into region_pmem.c
  dax/hmem, cxl: Defer and resolve Soft Reserved ownership
  cxl/region: Add helper to check Soft Reserved containment by CXL regions
  ...

dax/fsdev: fix uninitialized kaddr in fsdev_dax_zero_page_range()

2026-04-13T19:15:15+00:00

__fsdev_dax_direct_access() returns -EFAULT without setting *kaddr when
dax_pgoff_to_phys() returns -1 (pgoff out of range). The return value
was ignored, leaving kaddr uninitialized before being passed to
fsdev_write_dax().

Check the return value and propagate the error.

Thanks to Dan Carpenter and the smatch project for reporting this.

Signed-off-by: John Groves 
Reviewed-by: Jonathan Cameron 
Reviewed-by: Dave Jiang 
Link: https://patch.msgid.link/0100019d8262cda2-9714d31c-8fc1-4ca5-b32d-4df678240d14-000000@email.amazonses.com
Signed-off-by: Ira Weiny

mm: rename VMA flag helpers to be more readable

2026-04-05T20:53:18+00:00

Patch series "mm: vma flag tweaks".

The ongoing work around introducing non-system word VMA flags has
introduced a number of helper functions and macros to make life easier
when working with these flags and to make conversions from the legacy use
of VM_xxx flags more straightforward.

This series improves these to reduce confusion as to what they do and to
improve consistency and readability.

Firstly the series renames vma_flags_test() to vma_flags_test_any() to
make it abundantly clear that this function tests whether any of the flags
are set (as opposed to vma_flags_test_all()).

It then renames vma_desc_test_flags() to vma_desc_test_any() for the same
reason.  Note that we drop the 'flags' suffix here, as
vma_desc_test_any_flags() would be cumbersome and 'test' implies a flag
test.

Similarly, we rename vma_test_all_flags() to vma_test_all() for
consistency.

Next, we have a couple of instances (erofs, zonefs) where we are now
testing for vma_desc_test_any(desc, VMA_SHARED_BIT) &&
vma_desc_test_any(desc, VMA_MAYWRITE_BIT).

This is silly, so this series introduces vma_desc_test_all() so these
callers can instead invoke vma_desc_test_all(desc, VMA_SHARED_BIT,
VMA_MAYWRITE_BIT).

We then observe that quite a few instances of vma_flags_test_any() and
vma_desc_test_any() are in fact only testing against a single flag.

Using the _any() variant here is just confusing - 'any' of single item
reads strangely and is liable to cause confusion.

So in these instances the series reintroduces vma_flags_test() and
vma_desc_test() as helpers which test against a single flag.

The fact that vma_flags_t is a struct and that vma_flag_t utilises sparse
to avoid confusion with vm_flags_t makes it impossible for a user to
misuse these helpers without it getting flagged somewhere.

The series also updates __mk_vma_flags() and functions invoked by it to
explicitly mark them always inline to match expectation and to be
consistent with other VMA flag helpers.

It also renames vma_flag_set() to vma_flags_set_flag() (a function only
used by __mk_vma_flags()) to be consistent with other VMA flag helpers.

Finally it updates the VMA tests for each of these changes, and introduces
explicit tests for vma_flags_test() and vma_desc_test() to assert that
they behave as expected.


This patch (of 6):

On reflection, it's confusing to have vma_flags_test() and
vma_desc_test_flags() test whether any comma-separated VMA flag bit is
set, while also having vma_flags_test_all() and vma_test_all_flags()
separately test whether all flags are set.

Firstly, rename vma_flags_test() to vma_flags_test_any() to eliminate this
confusion.

Secondly, since the VMA descriptor flag functions are becoming rather
cumbersome, prefer vma_desc_test*() to vma_desc_test_flags*(), and also
rename vma_desc_test_flags() to vma_desc_test_any().

Finally, rename vma_test_all_flags() to vma_test_all() to keep the
VMA-specific helper consistent with the VMA descriptor naming convention
and to help avoid confusion vs.  vma_flags_test_all().

While we're here, also update whitespace to be consistent in helper
functions.

Link: https://lkml.kernel.org/r/cover.1772704455.git.ljs@kernel.org
Link: https://lkml.kernel.org/r/0f9cb3c511c478344fac0b3b3b0300bb95be95e9.1772704455.git.ljs@kernel.org
Signed-off-by: Lorenzo Stoakes (Oracle) 
Suggested-by: Pedro Falcato 
Acked-by: David Hildenbrand (Arm) 
Reviewed-by: Pedro Falcato 
Cc: Arnd Bergmann 
Cc: Babu Moger 
Cc: Baolin Wang 
Cc: Chao Yu 
Cc: Chatre, Reinette 
Cc: Chunhai Guo 
Cc: Damien Le Maol 
Cc: Dan Williams 
Cc: Dave Jiang 
Cc: Dave Martin 
Cc: Gao Xiang 
Cc: Greg Kroah-Hartman 
Cc: Hongbo Li 
Cc: Hugh Dickins 
Cc: James Morse 
Cc: Jan Kara 
Cc: Jann Horn 
Cc: Jason Gunthorpe 
Cc: Jeffle Xu 
Cc: Johannes Thumshirn 
Cc: Konstantin Komarov 
Cc: Liam Howlett 
Cc: "Luck, Tony" 
Cc: Matthew Wilcox (Oracle) 
Cc: Michal Hocko 
Cc: Mike Rapoport 
Cc: Muchun Song 
Cc: Naohiro Aota 
Cc: Oscar Salvador 
Cc: Sandeep Dhavale 
Cc: Suren Baghdasaryan 
Cc: Vishal Verma 
Cc: Vlastimil Babka 
Cc: Yue Hu 
Signed-off-by: Andrew Morton

dax/hmem: Parent dax_hmem devices

2026-04-01T15:12:18+00:00

For test purposes it is useful to be able to determine which
"hmem_platform" device is hosting a given sub-device.

Register hmem devices underneath "hmem_platform".

Signed-off-by: Dan Williams 
Reviewed-by: Ira Weiny 
Reviewed-by: Alison Schofield 
Reviewed-by: Dave Jiang 
Link: https://patch.msgid.link/20260327052821.440749-8-dan.j.williams@intel.com
Signed-off-by: Dave Jiang

dax/hmem: Fix singleton confusion between dax_hmem_work and hmem devices

2026-04-01T15:12:18+00:00

dax_hmem (ab)uses a platform device to allow for a module to autoload in
the presence of "Soft Reserved" resources. The dax_hmem driver had no
dependencies on the "hmem_platform" device being a singleton until the
recent "dax_hmem vs dax_cxl" takeover solution.

Replace the layering violation of dax_hmem_work assuming that there will
never be more than one "hmem_platform" device associated with a global work
item with a dax_hmem local workqueue that can theoretically support any
number of hmem_platform devices.

Fixup the reference counting to only pin the device while it is live in the
queue.

Signed-off-by: Dan Williams 
Reviewed-by: Ira Weiny 
Reviewed-by: Alison Schofield 
Reviewed-by: Dave Jiang 
Link: https://patch.msgid.link/20260327052821.440749-7-dan.j.williams@intel.com
Signed-off-by: Dave Jiang

dax/hmem: Reduce visibility of dax_cxl coordination symbols

2026-04-01T15:12:17+00:00

No other module or use case should be using dax_hmem_initial_probe or
dax_hmem_flush_work(). Limit their use to dax_hmem, and dax_cxl
respectively.

Signed-off-by: Dan Williams 
Reviewed-by: Ira Weiny 
Reviewed-by: Alison Schofield 
Reviewed-by: Dave Jiang 
Link: https://patch.msgid.link/20260327052821.440749-6-dan.j.williams@intel.com
Signed-off-by: Dave Jiang

cxl/region: Constify cxl_region_resource_contains()

2026-04-01T15:12:17+00:00

The call to cxl_region_resource_contains() in hmem_register_cxl_device()
need not cast away 'const'. The problem is the usage of the
bus_for_each_dev() API which does not mark its @data parameter as 'const'.
Switch to bus_find_device() which does take 'const' @data, fixup
cxl_region_resource_contains() and its caller.

Signed-off-by: Dan Williams 
Reviewed-by: Ira Weiny 
Reviewed-by: Alison Schofield 
Reviewed-by: Dave Jiang 
Link: https://patch.msgid.link/20260327052821.440749-5-dan.j.williams@intel.com
Signed-off-by: Dave Jiang

cxl/region: Limit visibility of cxl_region_contains_resource()

2026-04-01T15:12:17+00:00

The dax_hmem dependency on cxl_region_contains_resource() is a one-off
special case. It is not suitable for other use cases.

Move the definition to the other CONFIG_CXL_REGION guarded definitions in
drivers/cxl/cxl.h and include that by a relative path include. This matches
what drivers/dax/cxl.c does for its limited private usage of CXL core
symbols.

Reduce the symbol export visibility from global to just dax_hmem, to
further clarify its applicability.

Signed-off-by: Dan Williams 
Reviewed-by: Ira Weiny 
Reviewed-by: Alison Schofield 
Reviewed-by: Dave Jiang 
Link: https://patch.msgid.link/20260327052821.440749-4-dan.j.williams@intel.com
Signed-off-by: Dave Jiang

dax/cxl: Fix HMEM dependencies

2026-04-01T15:12:17+00:00

The expectation is that DEV_DAX_HMEM=y should be disallowed if any of
CXL_ACPI, or CXL_PCI are set =m. Also DEV_DAX_CXL=y should be disallowed if
DEV_DAX_HMEM=m. Use "$config || !$config" syntax for each dependency.
Otherwise, the invalid DEV_DAX_HMEM=m && DEV_DAX_CXL=y configuration is
allowed.

Lastly, dax_hmem depends on the availability of the
cxl_region_contains_resource() symbol published by the cxl_core.ko module.
So, also prevent DEV_DAX_HMEM from being built-in when the cxl_core module
is not built-in.

Signed-off-by: Dan Williams 
Reviewed-by: Ira Weiny 
Reviewed-by: Alison Schofield 
Reviewed-by: Dave Jiang 
Link: https://patch.msgid.link/20260327052821.440749-3-dan.j.williams@intel.com
Signed-off-by: Dave Jiang