Re: [PATCH v8 9/9] selftests: ftrace: Add wprobe trigger testcase

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The selftest incorrectly assumes that unlinking a file (`rm`) will synchronously trigger `dentry_kill()`. This only works on filesystems that do not cache negative dentries (like `tmpfs`). On standard disk filesystems (like `ext4`), the test will fail.
--

--- Patch [9]: [PATCH v8 9/9] selftests: ftrace: Add wprobe trigger testcase ---
commit 02c2b6c1500ba6cce360f9d5447d679e62801d61
Author: Masami Hiramatsu (Google) <mhiramat@kernel.org>

selftests: ftrace: Add wprobe trigger testcase

This commit adds a testcase for checking wprobe triggers. It sets set_wprobe
and clear_wprobe triggers on fprobe events to watch dentry access.

> diff --git a/tools/testing/selftests/ftrace/test.d/trigger/trigger-wprobe.tc b/tools/testing/selftests/ftrace/test.d/trigger/trigger-wprobe.tc
> new file mode 100644
> index 0000000000000..5ffe43f886d1a
> --- /dev/null
> +++ b/tools/testing/selftests/ftrace/test.d/trigger/trigger-wprobe.tc
[ ... ]
> +:;: "Start test workload" ;:
> +echo 1 >> tracing_on
> +
> +echo aaa > $TMPDIR/hoge
> +echo bbb > $TMPDIR/hoge
> +echo ccc > $TMPDIR/hoge
> +rm $TMPDIR/hoge
> +
> +:;: "Check trace results" ;:
> +cat trace | grep watch
> +
> +:;: "Ensure wprobe becomes disabled again" ;:
> +cat events/wprobes/watch/enable | grep 0

[Severity: Medium]
Will this regression cause the test to fail when $TMPDIR is on a standard disk
filesystem?

This selftest appears to rely on rm synchronously triggering dentry_kill() to
fire the clear_wprobe trigger.

While this works on memory-backed filesystems like tmpfs that do not cache
negative dentries (their d_delete operation immediately leads to dentry_kill()),
standard disk filesystems like ext4, xfs, or btrfs cache negative dentries to
optimize subsequent lookups. 

On those filesystems, unlinking the file leaves the dentry hashed on the dcache
LRU list, deferring dentry_kill() until memory pressure or a manual eviction
occurs. This would cause the test to fail when checking if the wprobe is
disabled.

Does this need to trigger a manual cache eviction or ensure the test runs on a
tmpfs mount to avoid failing?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/178417033089.209165.16717079876036408877.stgit@devnote2?part=9