DAMON development mailing list
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "SeongJae Park" <sj@kernel.org>
Cc: damon@lists.linux.dev
Subject: Re: [RFC PATCH v1.2 09/11] mm/damon/sysfs: kobject_del() probe filter dirs
Date: Thu, 18 Jun 2026 15:34:50 +0000	[thread overview]
Message-ID: <20260618153451.1E7171F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260618151517.5366-10-sj@kernel.org>

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

Pre-existing issues:
- [Medium] Missing `kobject_del()` call in `damos_sysfs_region_add_dirs()` error path causes delayed sysfs directory removal and subsequent `-EEXIST` name conflicts.
- [Medium] Missing `kobject_del()` in `damos_sysfs_probes_rm_dirs()` leaves child probe directories lingering in sysfs on error paths.
--

commit 6a62b56c2acdfd514de2a6e4cd36c3839e0c6db5
Author: SeongJae Park <sj@kernel.org>
Subject: mm/damon/sysfs: kobject_del() probe filter dirs

This commit fixes directory creation failures due to name conflicts on kernels
with CONFIG_DEBUG_KOBJECT_RELEASE. It does so by adding missing kobject_del()
calls before kobject_put() in the data attribute probe filter directories'
teardown path.

> diff --git a/mm/damon/sysfs.c b/mm/damon/sysfs.c
> index 5ca45654cdd39..36a0e82601697 100644
> --- a/mm/damon/sysfs.c
> +++ b/mm/damon/sysfs.c
> @@ -962,8 +962,10 @@ static void damon_sysfs_filters_rm_dirs(struct damon_sysfs_filters *filters)
>  	struct damon_sysfs_filter **filters_arr = filters->filters_arr;
>  	int i;
>  
> -	for (i = 0; i < filters->nr; i++)
> +	for (i = 0; i < filters->nr; i++) {
> +		kobject_del(&filters_arr[i]->kobj);
>  		kobject_put(&filters_arr[i]->kobj);
> +	}
>  	filters->nr = 0;
>  	kfree(filters_arr);
>  	filters->filters_arr = NULL;

[Severity: Medium]
This is a pre-existing issue, but since we are fixing missing kobject_del()
calls here, should we also update damos_sysfs_probes_rm_dirs() in
mm/damon/sysfs-schemes.c?

When tearing down probe directories, the loop iterates over child probes and
calls kobject_put(), but omitting kobject_del() might leave the successfully
added child directories in sysfs pending delayed release:

mm/damon/sysfs-schemes.c:damos_sysfs_probes_rm_dirs() {
    ...
	for (i = 0; i < probes->nr; i++)
		kobject_put(&probes_arr[i]->kobj);
    ...
}

Would this also result in -EEXIST name collisions on subsequent retry attempts?

[Severity: Medium]
This isn't a bug introduced by this patch, but does the error path in
damos_sysfs_region_add_dirs() need similar handling?

If kobject_init_and_add() succeeds but the subsequent call to
damos_sysfs_probes_add_dirs() fails, we jump to the fail path:

mm/damon/sysfs-schemes.c:damos_sysfs_region_add_dirs() {
    ...
	err = kobject_init_and_add(&probes->kobj, &damos_sysfs_probes_ktype,
			&region->kobj, "probes");
	if (err)
		goto fail;

	err = damos_sysfs_probes_add_dirs(probes, ctx, dregion);
	if (err)
		goto fail;
    ...
fail:
	kobject_put(&probes->kobj);
	return err;
}

Could the lack of kobject_del(&probes->kobj) here delay the removal of the
sysfs directory and cause an unexpected -EEXIST conflict if userspace
attempts to recreate the region?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260618151517.5366-1-sj@kernel.org?part=9

  reply	other threads:[~2026-06-18 15:34 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-18 15:15 [RFC PATCH v1.2 00/11] mm/damon/sysfs: kobject_del() directories that users can create/remove SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 01/11] mm/damon/sysfs: kobject_del() target (normal), context and kdamond dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 02/11] mm/damon/sysfs: kobject_del() region and target (error) dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 03/11] mm/damon/sysfs-schemes: kobject_del() scheme dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 04/11] mm/damon/sysfs-schemes: kobject_del() scheme region dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 05/11] mm/damon/sysfs-schemes: kobject_del() scheme filter dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 06/11] mm/damon/sysfs-schemes: kobject_del() scheme quota goal dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 07/11] mm/damon/sysfs-schemes: kobject_del() scheme action destination dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 08/11] mm/damon/sysfs: kobject_del() probe dirs SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 09/11] mm/damon/sysfs: kobject_del() probe filter dirs SeongJae Park
2026-06-18 15:34   ` sashiko-bot [this message]
2026-06-19  0:23     ` SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 10/11] mm/damon/sysfs: kobject_del() probe dirs in probes_addd_dir error path SeongJae Park
2026-06-18 15:15 ` [RFC PATCH v1.2 11/11] mm/damon/sysfs-schemes: kobject_del() region for populate_region error SeongJae Park
2026-06-18 15:38   ` sashiko-bot
2026-06-19  0:26     ` SeongJae Park
2026-06-19  0:30 ` [RFC PATCH v1.2 00/11] mm/damon/sysfs: kobject_del() directories that users can create/remove SeongJae Park

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260618153451.1E7171F000E9@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=damon@lists.linux.dev \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=sj@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox