On Fri, 5 Jun 2026 12:44:50 +0530 Sailesh Nandanavanam
<[email protected]> wrote:
> Hi SeongJae,
>
> Apologies for the delayed response.
No worry!
[...]
> I agree that a KUnit test is the right approach since it runs inside the
> kernel and can directly set walk_control_obsolete to simulate the race
> condition reliably without timing dependencies.
>
> I will drop the userspace selftest approach and write a KUnit test for
> damos_walk() functionality including walk_control_obsolete. I will verify
> it fails on a kernel with commit 33c3f6c2b48c ("mm/damon/core: fix
> damos_walk() vs kdamond_fn() exit race") reverted and passes with the fix
> present before sending v3.
>
> Regarding other feedback:
> - Will use correct commit description format with subject in brackets
> - Will remove the Fixes: tag
> - Will add blank line before _damon_sysfs import
> - Will add changelog and link previous versions
> - Will share revision plan and wait at least one day
> before sending new version
Sounds good. Please consider using in-line reply [1] from the next time,
though!
>
> Please give me some time to study mm/damon/tests/core-kunit.h and implement
> this properly.
No rush, take your time.
[1] https://subspace.kernel.org/etiquette.html#do-not-top-post-when-replying
Thanks,
SJ
[...]
