[PATCH v2 0/2] gpu/buddy: Fix quadratic cost of range-biased allocation

[Sanjay Yadav <sanjay.kumar.yadav@intel.com>]

Range-restricted allocations (GPU_BUDDY_RANGE_ALLOCATION) go through
__alloc_range_bias(), a depth-first walk over the buddy tree that visits
the higher-address child first. It skips blocks that are too small,
outside the range or ALLOCATED, but always descends SPLIT blocks, even
when nothing below them is free. Range allocations are placed top-down,
so they pack below the range end and each new allocation walks over all
blocks already allocated there: O(N) per allocation, O(N^2) in total,
under the driver's allocator lock.

This is a property of the range-biased allocation algorithm rather than
of any particular memory configuration. Any driver that repeatedly
restricts placement to a window can encounter this behavior once enough
small allocations accumulate in that window.

One configuration where the problem becomes particularly visible is a
small-BAR system. Drivers restrict every CPU-visible allocation to that
window with a range placement (fpfn = 0, lpfn = BAR size).

xe_pat@pt-caching-random-offsets does about 130k 4K range allocations
and hit the CI 180s inactivity timeout on a debug kernel inside
gpu_buddy_alloc_blocks(). With the full VRAM visible the same test does
not use the range path.

<6> [59.426526] [IGT] xe_pat: executing
<6> [59.436001] [IGT] xe_pat: starting subtest pt-caching-random-offsets
<6> [239.585293] task:xe_pat          state:R  running task     stack:0
pid:3700  tgid:3700  ppid:1957   task_flags:0x400100 flags:0x00000020
<6> [239.585304] Call Trace:
<6> [239.585307]  <TASK>
<6> [239.585309]  ? irqentry_exit+0x2a2/0x8d0
<6> [239.585314]  ? trace_hardirqs_off_finish+0xc7/0xe0
<6> [239.585320]  ? sysvec_apic_timer_interrupt+0x57/0xc0
<6> [239.585326]  ? asm_sysvec_apic_timer_interrupt+0x1b/0x20
<6> [239.585334]  ? gpu_buddy_alloc_blocks+0x442/0xec0
<6> [239.585341]  ? gpu_buddy_alloc_blocks+0xa37/0xec0
<6> [239.585353]  ? xe_ttm_vram_mgr_new+0x1c7/0x340 [xe]
<6> [239.585506]  ? ttm_resource_alloc+0x50/0xd0 [ttm]
<6> [239.585518]  ? ttm_bo_alloc_at_place+0x84/0x180 [ttm]
<6> [239.585527]  ? ttm_bo_alloc_resource+0x115/0x5e0 [ttm]
<6> [239.585538]  ? ttm_bo_validate+0x94/0x1c0 [ttm]
<6> [239.585546]  ? lock_is_held_type+0xa3/0x130
<6> [239.585552]  ? ttm_bo_init_reserved+0xa5/0x1f0 [ttm]
<6> [239.585560]  ? xe_bo_init_locked+0x20a/0x650 [xe]
<6> [239.585616]  ? __pfx_xe_ttm_bo_destroy+0x10/0x10 [xe]
<6> [239.585672]  ? stack_depot_save_flags+0x28/0x960
<6> [239.585679]  ? __xe_bo_create_locked+0xb1/0xa20 [xe]
<6> [239.585733]  ? _raw_spin_unlock_irqrestore+0x52/0x80
<6> [239.585743]  ? __create_object+0x67/0xe0
<6> [239.585747]  ? trace_hardirqs_on+0x22/0xe0
<6> [239.585753]  ? _raw_spin_unlock_irqrestore+0x52/0x80
<6> [239.585760]  ? xe_bo_create_pin_map_at_aligned+0x51/0x1f0 [xe]
<6> [239.585813]  ? __kmalloc_cache_noprof+0x494/0x670
<6> [239.585819]  ? xe_bo_create_pin_map+0x1f/0x40 [xe]
<6> [239.585870]  ? xe_pt_create+0xb5/0x430 [xe]

Patch 1 tracks, per block, the largest order of any FREE block in its
subtree and lets __alloc_range_bias() skip subtrees that cannot satisfy
the request. Skipped subtrees contain nothing the old walk could split
or return, and the remaining blocks are visited in the same order, so
the selected block and placement are unchanged. The new s8 field uses
existing tail padding.

Patch 2 adds KUnit tests. One fills a range prefix the way small-BAR
drivers do and checks pruning deterministically by hiding a free block
behind an ancestor that reports no free space. The other checks the
metadata of every block after each state change, including coalescing,
rollback of a partial allocation, and trim.

Results
  n_alloc     avg_blocks_visited        avg_allocation_time
              without fix  with fix     without fix  with fix
     4096          4141        46          10.7 us    0.293 us
    65536        127041        48           536 us    0.259 us
   126976        249911        47          1.46 ms    0.283 us

v2:
- Reword cover letter and to describe this as a general fix for the
  algorithmic cost of range-biased allocation rather than a small-BAR
  specific issue (Matt A)
- No code changes

Sanjay Yadav (2):
  gpu/buddy: Prune exhausted subtrees in range-biased allocation
  gpu/tests/buddy: Add range-bias pruning and free-order tracking tests

 drivers/gpu/buddy.c                |  31 +++
 drivers/gpu/tests/gpu_buddy_test.c | 307 +++++++++++++++++++++++++++++
 include/linux/gpu_buddy.h          |   7 +
 3 files changed, 345 insertions(+)

-- 
2.52.0