linux-stable.git/tools/testing/memblock, branch master

Merge tag 'liveupdate-v7.3-rc1-20260823' of git://git.kernel.org/pub/scm/linux/kernel/git/liveupdate/linux

2026-08-23T16:17:38+00:00

Pull more liveupdate updates from Mike Rapoport:
 "Make boot time huge page allocation work nicely with kexec handover.

  Today allocation of gigantic pages in HugeTLB cannot work reliably
  with kexec handover (KHO):

   - HugeTLB allocates gigantic pages using memblock and autoscaling of
     KHO scratch accounts for these allocations. When gigantic pages
     occupy half of the memory of more, KHO fails to allocate its
     scratch memory.

   - After kexec handover, memblock allocations exclusively use KHO
     scratch that is not supposed to contain preserved memory. This
     essentially blocks preservation of HugeTLB with gigantic pages.

  Extend early memory pools available for KHO kernel with areas that are
  guaranteed not to contain preserved memory"

* tag 'liveupdate-v7.3-rc1-20260823' of git://git.kernel.org/pub/scm/linux/kernel/git/liveupdate/linux: (21 commits)
  kho: exclude hugetlb memory from scratch size calculation
  memblock: add memblock_reserved_hugetlb_size()
  memblock: make HugeTLB bootmem allocation work with KHO
  memblock: always include KHO headers
  kho: extend scratch
  mm/mm_init: don't rely on memblock to get KHO scratch migratetype
  kho: initialize preserved memory map radix tree earlier
  kho: initialize kho_scratch pointer earlier in boot
  kho: expose kho_scratch_overlap() to kexec_handover.h
  kho: add kho_radix_init_tree()
  kho: allow destroying KHO radix tree
  kho: allow early-boot usage of the KHO radix tree
  kho: add data argument to radix walk callback
  kho: add callback for table pages
  kho: add a struct for radix callbacks
  kho: move all memory retrieval logic to kho_mem_retrieve()
  kho: store incoming radix tree in kho_in
  kho: disallow wide keys in radix tree
  kho: make radix max key width more obvious
  kho: generalize radix tree APIs
  ...

mm: split out mm_init and memblock declarations from internal.h

2026-08-05T02:18:45+00:00

Patch series "mm: split a couple of headers from internal.h", v2.

mm/internal.h becomes more and more bloated.

Split declarations related to mm_init, memblock, vmalloc and sparse into
new headers.


This patch (of 3):

mm/internal.h becomes more and more bloated.

Move declarations for related to mm/mm_init.c and mm/memblock.c to a new
mm/mm_init.h header.

No functional changes.

[rppt@kernel.org: split stubfs from internal.h to mm_init.h]
  Link: https://lore.kernel.org/alJd1BLypyK9Mpaw@kernel.org
Link: https://lore.kernel.org/20260709-internal-h-v2-0-695631425968@kernel.org
Link: https://lore.kernel.org/20260709-internal-h-v2-1-695631425968@kernel.org
Signed-off-by: Mike Rapoport (Microsoft) 
Acked-by: Muchun Song 
Acked-by: Vlastimil Babka (SUSE) 
Acked-by: David Hildenbrand (Arm) 
Acked-by: Lorenzo Stoakes 
Acked-by: Pratyush Yadav 
Acked-by: SJ Park 
Cc: Alexander Graf 
Cc: Alexander Potapenko 
Cc: Brendan Jackman 
Cc: Brendan Jackman 
Cc: Dennis Zhou 
Cc: Dmitry Vyukov 
Cc: Johannes Weiner 
Cc: Liam R. Howlett 
Cc: Marco Elver 
Cc: Michal Hocko 
Cc: Oscar Salvador 
Cc: Pasha Tatashin 
Cc: Suren Baghdasaryan 
Cc: Tejun Heo 
Cc: "Uladzislau Rezki (Sony)" 
Cc: Zi Yan 
Signed-off-by: Andrew Morton

memblock: always include KHO headers

2026-08-04T06:28:18+00:00

In a coming commit, memblock will start using kho_scratch_overlap()
without a compile guard. The compile guard for the function is in
kexec_handover.h and provides a stub when CONFIG_KEXEC_HANDOVER is
disabled.

Since in memblock the call will exist unconditionally, always include
the KHO headers.

Including these headers unconditionally breaks memblock test
compilation. Add stubs to fix that.

Signed-off-by: Pratyush Yadav (Google) 
Link: https://patch.msgid.link/20260801084833.1897543-20-pratyush@kernel.org
Signed-off-by: Mike Rapoport (Microsoft)

kho: expose kho_scratch_overlap() to kexec_handover.h

2026-08-04T06:23:05+00:00

Support for discovering memory blocks with no preserved memory will be
added in coming patches. These areas will also be marked as scratch to
allow allocations from them. Memblock will switch to looking through the
scratch array to decide the right migratetype.

Expose kho_scratch_overlap() to KHO users. Since it is now used by
non-debug code, move it out of kexec_handover_debug.c and into
kexec_handover.c. Gate the overlap checks in kho_preserve_folio() and
kho_preserve_pages() by IS_ENABLED(CONFIG_KEXEC_HANDOVER_DEBUG) instead.

Since kexec_handover_debug.c is now empty, delete it.

Add a stub for kho_scratch_overlap() to memblock tests to make sure it
compiles. It will be used in memblock by a coming commit.

No functional changes.

Signed-off-by: Pratyush Yadav (Google) 
Link: https://patch.msgid.link/20260801084833.1897543-14-pratyush@kernel.org
Signed-off-by: Mike Rapoport (Microsoft)

Merge tag 'liveupdate-v7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/liveupdate/linux

2026-06-21T16:46:14+00:00

Pull liveupdate updates from Mike Rapoport:
 "Kexec Handover (KHO):

   - make memory preservation compatible with deferred initialization
     of the memory map

  Live Update Orchestrator (LUO):

   - add LIVEUPDATE_SESSION_GET_NAME ioctl and parameter verification
     for LIVEUPDATE_IOCTL_CREATE_SESSION ioctl

   - documentation updates for liveupdate=on command line option,
     systemd support and the current compatibility status

   - remove the fixed limits on the number of files that can be
     preserved within a single session, and the total number of
     sessions managed by the LUO

  Misc fixes:

   - reference count incoming File-Lifecycle-Bound (FLB) data so
     it cannot be freed while a subsystem is still using it

   - fixes for a TOCTOU race in luo_session_retrieve(), a use-
     after-free in the file finish and unpreserve paths, concurrent
     session mutations during reboot and serialization on
     preserve_context kexec

   - make sure ioctls for incoming LUO sessions are blocked for
     outgoing sessions and vice versa

   - make sure KHO scratch size is always aligned by
     CMA_MIN_ALIGNMENT_BYTES

   - fix memblock tests build issue introduced by KHO changes"

* tag 'liveupdate-v7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/liveupdate/linux: (36 commits)
  liveupdate: Document that retrieve failure is permanent
  docs: memfd_preservation: fix rendering of ABI documentation
  selftests/liveupdate: Add stress-files kexec test
  selftests/liveupdate: Add stress-sessions kexec test
  selftests/liveupdate: Test session and file limit removal
  liveupdate: Remove limit on the number of files per session
  liveupdate: Remove limit on the number of sessions
  liveupdate: defer session block allocation and physical address setting
  kho: add support for linked-block serialization
  liveupdate: Extract luo_session_deserialize_one helper
  liveupdate: Extract luo_file_deserialize_one helper
  liveupdate: register luo_ser as KHO subtree
  liveupdate: centralize state management into struct luo_ser
  liveupdate: avoid mixing cleanup guards with goto in luo_session_retrieve_fd
  liveupdate: change file_set->count type to u64 for type safety
  liveupdate: Remove unused ser field from struct luo_session
  liveupdate: fix u-a-f in luo_file_unpreserve_files() and luo_file_finish()
  liveupdate: block session mutations during reboot
  liveupdate: fix TOCTOU race in luo_session_retrieve()
  liveupdate: skip serialization for context-preserving kexec
  ...

tools/testing/memblock: fix stale NUMA reservation tests

2026-06-02T05:34:03+00:00

memblock allocations now reserve memory with MEMBLOCK_RSRV_KERN and,
on NUMA configurations, record the requested node on the reserved
region. Several memblock simulator NUMA tests still expected merges
that only worked before those reservation semantics changed, so the
suite aborted even though the allocator behavior was correct.

Update the NUMA merge expectations in the memblock_alloc_try_nid()
and memblock_alloc_exact_nid_raw() tests to match the current reserved
region metadata rules. For cases that should still merge, create the
pre-existing reservation with matching nid and MEMBLOCK_RSRV_KERN
metadata. Also strengthen the memblock_alloc_node() coverage by
checking the newly created reserved region directly instead of
re-reading the source memory node descriptor.

Finally, drop the stale README/TODO notes that still claimed
memblock_alloc_node() could not be tested.

The memblock simulator passes again with NUMA enabled after these
updates.

Signed-off-by: Priyanshu Kumar 
Link: https://patch.msgid.link/20260415122731.1768912-1-priyanshukumarpu@gmail.com
[rppt: dropped unrelated changes]
Signed-off-by: Mike Rapoport (Microsoft)

memblock tests: define MIGRATE_CMA

2026-05-31T23:31:38+00:00

kho_scratch_migratetype(), defined in include/linux/memblock.h uses enum
migratetype. This breaks build for memblock tests with:

./linux/memblock.h:634:73: error: parameter 2 (‘mt’) has incomplete type
  634 |                                                        enum migratetype mt)

Fix it by defining enum migratetype and MIGRATE_CMA. As is the case with
the other headers in tools/testing/memblock, do not bring in the whole
thing, only what is needed.

Reported-by: Mike Rapoport 
Closes: https://lore.kernel.org/linux-mm/afcdDm4aAJvNaQqH@kernel.org/
Signed-off-by: Pratyush Yadav (Google) 
Link: https://patch.msgid.link/20260504102742.3833159-1-pratyush@kernel.org
Signed-off-by: Mike Rapoport (Microsoft)

mm: move free_reserved_area() to mm/memblock.c

2026-04-01T08:19:46+00:00

free_reserved_area() is related to memblock as it frees reserved memory
back to the buddy allocator, similar to what memblock_free_late() does.

Move free_reserved_area() to mm/memblock.c to prepare for further
consolidation of the functions that free reserved memory.

No functional changes.

Link: https://patch.msgid.link/20260323074836.3653702-5-rppt@kernel.org
Signed-off-by: Mike Rapoport (Microsoft) 
Acked-by: Vlastimil Babka (SUSE)

memblock: move reserve_bootmem_range() to memblock.c and make it static

2026-04-01T08:19:45+00:00

reserve_bootmem_region() is only called from
memmap_init_reserved_pages() and it was in mm/mm_init.c because of its
dependecies on static init_deferred_page().

Since init_deferred_page() is not static anymore, move
reserve_bootmem_region(), rename it to memmap_init_reserved_range() and
make it static.

Update the comment describing it to better reflect what the function
does and drop bogus comment about reserved pages in free_bootmem_page().

Update memblock test stubs to reflect the core changes.

Reviewed-by: Lorenzo Stoakes (Oracle) 
Reviewed-by: David Hildenbrand (Arm) 
Link: https://patch.msgid.link/20260323072042.3651061-1-rppt@kernel.org
Signed-off-by: Mike Rapoport (Microsoft)

memblock: Add reserve_mem debugfs info

2026-04-01T08:19:43+00:00

When using the "reserve_mem" parameter, users aim at having an
area that (hopefully) persists across boots, so pstore infrastructure
(like ramoops module) can make use of that to save oops/ftrace logs,
for example.

There is no easy way to determine if this kernel parameter is properly
set though; the kernel doesn't show information about this memory in
memblock debugfs, neither in /proc/iomem nor dmesg. This is a relevant
information for tools like kdumpst[0], to determine if it's reliable
to use the reserved area as ramoops persistent storage; checking only
/proc/cmdline is not sufficient as it doesn't tell if the reservation
effectively succeeded or not.

Add here a new file under memblock debugfs showing properly set memory
reservations, with name and size as passed to "reserve_mem". Notice that
if no "reserve_mem=" is passed on command-line or if the reservation
attempts fail, the file is not created.

[0] https://aur.archlinux.org/packages/kdumpst

Reviewed-by: SeongJae Park 
Signed-off-by: Guilherme G. Piccoli 
Link: https://patch.msgid.link/20260324012839.1991765-2-gpiccoli@igalia.com
Signed-off-by: Mike Rapoport (Microsoft)