Test bots don't seem to test kmalloc_nolock() and kfree_nolock(), so
subtle bugs are discovered late. To make it easier to test, add a kunit
test case that stresses these paths by combining a tight alloc/free loop
with perf-triggered interrupts that invoke {kmalloc,kfree}_nolock() from
the overflow handler.

While at it, also add the missing MAINTAINERS entry for slub_kunit.c.

v1->v2:
  - Fix build errors on arches without CONFIG_PERF_EVENTS (Thanks
    Vlastimil)
  - Rename some identifiers
  - Use GFP_KERNEL_ACCOUNT or GFP_KERNEL for kmalloc(), instead of 0 or
    __GFP_ACCOUNT (that was unnecessarily restrictive).

p.s. I was writing this test while developing kvfree_rcu_nolock(),
let's add a test case in advance.

Probably a similar test case for the page allocator is worth adding
but I'm not sure where the right place would be.

v1: https://lore.kernel.org/linux-mm/20260330120517.104743-1-harry@kernel.org

Harry Yoo (Oracle) (2):
  MAINTAINERS: add lib/tests/slub_kunit.c to SLAB ALLOCATOR section
  lib/tests/slub_kunit: add a test case for {kmalloc,kfree}_nolock

 MAINTAINERS            |  1 +
 lib/tests/slub_kunit.c | 92 ++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 93 insertions(+)


base-commit: dbc785fb4d9cfa0115e69e614fe20b4ab90f8366
-- 
2.43.0