| Age | Commit message (Collapse) | Author |
|---|
|
# Conflicts:
# drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
# drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
|
|
The try_harder contiguous fallback could return a range whose start
offset did not match the caller's min_block_size. When a candidate's
start is misaligned, realign it: free the misaligned run and reallocate
exactly @size at the next lower min_block_size boundary. This keeps the
returned size unchanged with no surplus to trim, and rejects the request
only when no aligned candidate fits.
v2: align misaligned candidates down to min_block_size instead of
bailing out, for both the RHS and LHS paths (Matthew).
Fixes: 0a1844bf0b53 ("drm/buddy: Improve contiguous memory allocation")
Suggested-by: Christian König <christian.koenig@amd.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Timur Kristóf <timur.kristof@gmail.com>
Cc: stable@vger.kernel.org
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Tested-by: John Olender <john.olender@gmail.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Link: https://patch.msgid.link/20260709131050.1022759-1-Arunpravin.PaneerSelvam@amd.com
|
|
Extend the scoreboard approach from the previous commit to used blocks,
so drm_buddy_print() can report per-order allocation pressure in O(1).
Unlike free blocks, an allocated block can leave the allocated state
through mark_free() (normal free and gpu_buddy_block_trim()) or be
consumed directly by gpu_block_free() during coalescing. Both sites are
guarded by gpu_buddy_block_is_allocated() and paired with the increment
in mark_allocated().
v3:
- Assert scoreboard is empty at fini(), as sanity check (Matthew Auld)
v2:
- Update after fix for use-after-free in split_block() call sites
- Change goto label to out_free_used_scoreboard for clarity
- Make drm_buddy_print() and gpu_buddy_print() symmetric for used and
free
Assisted-by: GitHub Copilot:claude-sonnet-4.6
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20260522092600.32818-6-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
|
|
Reporting per-order free block counts in drm_buddy_print() currently
requires walking all rbtrees, which is O(n) over the total number of
free blocks and holds the allocator lock for the duration. This becomes
expensive on large VRAM heaps with many small free fragments.
Maintain a free_scoreboard[] array indexed by order instead, so that
the count for any order is always available in O(1). The scoreboard is
kept accurate by hooking into the four places where a block's free state
changes: mark_free(), mark_allocated(), mark_split(), and the sites in
__gpu_buddy_free(), __force_merge(), and the four err_undo paths that
call rbtree_remove() directly on free blocks without going through
mark_*().
The print functions are simplified as a result: the rbtree traversal
is replaced by a direct array lookup.
v3: Update after introducing __gpu_buddy_undo_splits() helper
v2: Update after fix for use-after-free in split_block() call sites
Assisted-by: GitHub Copilot:claude-sonnet-4.6
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20260522092600.32818-5-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
|
|
The pattern of merging a block back with its buddy on error paths is
duplicated across multiple locations. Extract it into a
__gpu_buddy_undo_splits() helper to avoid repetition and prepare for
future changes.
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20260522092600.32818-4-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
|
|
The err_undo label in alloc_from_freetree() is only reachable via a
goto from inside the `while (tmp != order)` loop, which means tmp is
guaranteed to differ from order at that point. The surrounding
`if (tmp != order)` guard was therefore always true and can be dropped
without any behavioral change.
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20260522092600.32818-3-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
|
|
When split_block() fails it returns before calling mark_split(), leaving
the block in the FREE state and still linked in the rbtree. The four
err_undo paths then call __gpu_buddy_free() without first removing the
block from the tree, which leads to two distinct bugs:
- If the buddy is also free, __gpu_buddy_free() merges the two siblings
by calling gpu_block_free(mm, block) while block->rb is still linked
in the tree. Any subsequent rbtree traversal will follow the now-
dangling pointer, causing a use-after-free.
- In alloc_from_freetree(), where there is no buddy guard,
__gpu_buddy_free() always reaches mark_free() -> rbtree_insert() with
block still in the tree, corrupting the rbtree.
The same pattern is already used correctly in __force_merge(): call
rbtree_remove() to unlink the block before handing it to
__gpu_buddy_free(). Apply the same fix to all four err_undo sites.
Reported-by: Sashiko <sashiko-bot@kernel.org>
Assisted-by: GitHub Copilot:claude-sonnet-4.6
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20260522092600.32818-2-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
|
|
gpu_buddy APIs are expected to be called with the driver-provided lock
held, but there is no runtime enforcement of this contract. Add lockdep
annotations to catch locking violations early.
Introduce gpu_buddy_driver_set_lock() for the driver to register the
lock that protects the buddy manager. Add gpu_buddy_driver_lock_held()
assertions to all exported gpu_buddy and drm_buddy APIs that
access/modify the manager state. The lock_dep_map field is only compiled
in when CONFIG_LOCKDEP is enabled, adding zero overhead to production
builds.
Wire up xe_ttm_vram_mgr to register its mutex with the buddy manager
after initialization.
Assisted-by: Copilot:claude-opus-4.6
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Link: https://patch.msgid.link/20260508065544.4049240-2-tejas.upadhyay@intel.com
|
|
Large alignment requests previously forced the buddy allocator to search by
alignment order, which often caused higher-order free blocks to be split even
when a suitably aligned smaller region already existed within them. This led
to excessive fragmentation, especially for workloads requesting small sizes
with large alignment constraints.
This change prioritizes the requested allocation size during the search and
uses an augmented RB-tree field (subtree_max_alignment) to efficiently locate
free blocks that satisfy both size and offset-alignment requirements. As a
result, the allocator can directly select an aligned sub-region without
splitting larger blocks unnecessarily.
A practical example is the VKCTS test
dEQP-VK.memory.allocation.basic.size_8KiB.reverse.count_4000, which repeatedly
allocates 8 KiB buffers with a 256 KiB alignment. Previously, such allocations
caused large blocks to be split aggressively, despite smaller aligned regions
being sufficient. With this change, those aligned regions are reused directly,
significantly reducing fragmentation.
This improvement is visible in the amdgpu VRAM buddy allocator state
(/sys/kernel/debug/dri/1/amdgpu_vram_mm). After the change, higher-order blocks
are preserved and the number of low-order fragments is substantially reduced.
Before:
order- 5 free: 1936 MiB, blocks: 15490
order- 4 free: 967 MiB, blocks: 15486
order- 3 free: 483 MiB, blocks: 15485
order- 2 free: 241 MiB, blocks: 15486
order- 1 free: 241 MiB, blocks: 30948
After:
order- 5 free: 493 MiB, blocks: 3941
order- 4 free: 246 MiB, blocks: 3943
order- 3 free: 123 MiB, blocks: 4101
order- 2 free: 61 MiB, blocks: 4101
order- 1 free: 61 MiB, blocks: 8018
By avoiding unnecessary splits, this change improves allocator efficiency and
helps maintain larger contiguous free regions under heavy offset-aligned
allocation workloads.
v2:(Matthew)
- Update augmented information along the path to the inserted node.
v3:
- Move the patch to gpu/buddy.c file.
v4:(Matthew)
- Use the helper instead of calling _ffs directly
- Remove gpu_buddy_block_order(block) >= order check and drop order
- Drop !node check as all callers handle this already
- Return larger than any other possible alignment for __ffs64(0)
- Replace __ffs with __ffs64
v5:(Matthew)
- Drop subtree_max_alignment initialization at gpu_block_alloc()
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patch.msgid.link/20260306060155.2114-1-Arunpravin.PaneerSelvam@amd.com
|
|
Introduce gpu_buddy_assert(), a small helper that wraps WARN_ON() and,
when CONFIG_KUNIT is enabled, also calls kunit_fail_current_test() so
that any active KUnit test is marked as failed. In non-KUnit builds
the macro reduces to WARN_ON(), preserving existing behaviour.
Stringify the asserted condition in the failure message to make it easy
to identify which assertion fired. Leave the WARN_ON() in
gpu_buddy_block_trim() unchanged, as it returns -EINVAL and the caller
already observes the failure via the return code.
Cc: Christian König <christian.koenig@amd.com>
Cc: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Sanjay Yadav <sanjay.kumar.yadav@intel.com>
Reviewed-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Link: https://patch.msgid.link/20260227130037.53615-2-sanjay.kumar.yadav@intel.com
|
|
Move gpu_buddy_block_state(), gpu_buddy_block_is_allocated(),
and gpu_buddy_block_is_split() from gpu_buddy.h to gpu_buddy.c
as static functions since they have no external callers.
Remove gpu_get_buddy() as it was an unused exported wrapper
around the internal __get_buddy().
No functional changes.
v2:
- Rebased after DRM buddy allocator moved to drivers/gpu/
- Keep gpu_buddy_block_is_free() in header since it's now
used by drm_buddy.c
- Updated commit message
Cc: Christian König <christian.koenig@amd.com>
Cc: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Sanjay Yadav <sanjay.kumar.yadav@intel.com>
Reviewed-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Link: https://patch.msgid.link/20260212092527.718455-6-sanjay.kumar.yadav@intel.com
|
|
Move the DRM buddy allocator one level up so that it can be used by GPU
drivers (example, nova-core) that have usecases other than DRM (such as
VFIO vGPU support). Modify the API, structures and Kconfigs to use
"gpu_buddy" terminology. Adapt the drivers and tests to use the new API.
The commit cannot be split due to bisectability, however no functional
change is intended. Verified by running K-UNIT tests and build tested
various configurations.
Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
[airlied: I've split this into two so git can find copies easier.
I've also just nuked drm_random library, that stuff needs to be done
elsewhere and only the buddy tests seem to be using it].
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Move the DRM buddy allocator one level up so that it can be used by GPU
drivers (example, nova-core) that have usecases other than DRM (such as
VFIO vGPU support). Modify the API, structures and Kconfigs to use
"gpu_buddy" terminology. Adapt the drivers and tests to use the new API.
The commit cannot be split due to bisectability, however no functional
change is intended. Verified by running K-UNIT tests and build tested
various configurations.
Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
[airlied: I've split this into two so git can find copies easier.
I've also just nuked drm_random library, that stuff needs to be done
elsewhere and only the buddy tests seem to be using it].
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
