BPF List
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Ihor Solodrai" <ihor.solodrai@linux.dev>
Cc: bpf@vger.kernel.org
Subject: Re: [PATCH bpf-next v1 2/4] libbpf: Render decl_tags in btf_dump
Date: Thu, 17 Sep 2026 01:30:48 +0000	[thread overview]
Message-ID: <20260917013049.2263A1F000FF@smtp.kernel.org> (raw)
In-Reply-To: <20260917012037.1396254-3-ihor.solodrai@linux.dev>

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] libbpf: btf_dump_cmp_decl_tags violates strict weak ordering in qsort
- [Medium] libbpf: duplicate tags and memory leak on btf_dump_resize error paths
--

commit 869ac84b9144d9ee48198dbc0d57c9510a6f9d98
Author: Ihor Solodrai <ihor.solodrai@linux.dev>
Subject: libbpf: Render decl_tags in btf_dump

This commit adds support for rendering BTF_KIND_DECL_TAG in btf_dump.
It builds an array of decl tags sorted by target ID and tag ID, which
are then emitted at every declaration.

> diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
> --- a/tools/lib/bpf/btf_dump.c
> +++ b/tools/lib/bpf/btf_dump.c
> @@ -202,9 +202,15 @@ struct btf_dump *btf_dump__new(const struct btf *btf,
> +static int btf_dump_cmp_decl_tags(const void *a, const void *b)
> +{
> +	const struct decl_tag_desc *x = a, *y = b;
> +
> +	if (x->target_id != y->target_id)
> +		return x->target_id < y->target_id ? -1 : 1;
> +	return x->tag_id < y->tag_id ? -1 : 1;

[Severity: Medium]
Does this comparator satisfy the strict weak ordering requirement for qsort?

If two elements have the same tag_id, this returns 1 instead of 0. Depending
on the libc implementation, returning 1 when comparing an element to itself
or to a duplicate could break antisymmetry and potentially lead to out of
bounds accesses or infinite loops.

> +}

[ ... ]

> @@ -258,8 +258,18 @@ static int btf_dump_resize(struct btf_dump *d)
>  		err = btf_dump_mark_referenced(d, t);
>  		if (err)
>  			return err;

[Severity: Medium]
Do we need to roll back d->decl_tags_cnt or free resources here?

If this loop encounters an error and returns early, it leaves the previously
pushed decl_tags in the array, but d->last_id isn't updated.

If a user handles the error and resumes using the btf_dump object, won't it
re-push those same tags on the next call, causing duplicates and memory leaks?

> +
> +		if (btf_is_decl_tag(t)) {
> +			err = btf_dump_push_decl_tag(d, i, t);
> +			if (err)
> +				return err;

[Severity: Medium]
Similarly, does returning early here lead to the same duplicate tag entries
on retry, since d->decl_tags_cnt is not rolled back on error?

> +		}
>  	}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260917012037.1396254-1-ihor.solodrai@linux.dev?part=2

  reply	other threads:[~2026-09-17  1:30 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-17  1:20 [PATCH bpf-next v1 0/4] libbpf: Render decl_tags in btf_dump Ihor Solodrai
2026-09-17  1:20 ` [PATCH bpf-next v1 1/4] libbpf: Walk types in btf_dump_resize(), not in mark_referenced() Ihor Solodrai
2026-09-17 22:32   ` Eduard Zingerman
2026-09-17  1:20 ` [PATCH bpf-next v1 2/4] libbpf: Render decl_tags in btf_dump Ihor Solodrai
2026-09-17  1:30   ` sashiko-bot [this message]
2026-09-17  2:12   ` Alexei Starovoitov
2026-09-17  3:25     ` Ihor Solodrai
2026-09-17 22:46   ` Eduard Zingerman
2026-09-17 23:59   ` Andrii Nakryiko
2026-09-22 16:59     ` Ihor Solodrai
2026-09-22 19:43       ` Andrii Nakryiko
2026-09-17  1:20 ` [PATCH bpf-next v1 3/4] selftests/bpf: Test btf_dump rendering of decl_tags Ihor Solodrai
2026-09-17 10:30   ` Alan Maguire
2026-09-17 17:38     ` Ihor Solodrai
2026-09-17 23:18   ` Eduard Zingerman
2026-09-17  1:20 ` [PATCH bpf-next v1 4/4] selftests/bpf: Show the tag kflag in the raw BTF dump helper Ihor Solodrai
2026-09-17  1:27   ` sashiko-bot
2026-09-17 23:21   ` Eduard Zingerman

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=20260917013049.2263A1F000FF@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=ihor.solodrai@linux.dev \
    --cc=sashiko-reviews@lists.linux.dev \
    /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