From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
To: "Seokwoo Chung (Ryan)" <seokwoo.chung130@gmail.com>
Cc: rostedt@goodmis.org, corbet@lwn.net, shuah@kernel.org,
mathieu.desnoyers@efficios.com, linux-kernel@vger.kernel.org,
linux-trace-kernel@vger.kernel.org, linux-doc@vger.kernel.org,
linux-kselftest@vger.kernel.org
Subject: Re: [PATCH v6 2/4] fprobe: Support comma-separated filters in register_fprobe()
Date: Tue, 24 Mar 2026 10:59:13 +0900 [thread overview]
Message-ID: <20260324105913.cf60a2f2fcc64afbe30bd22e@kernel.org> (raw)
In-Reply-To: <20260205135842.20517-3-seokwoo.chung130@gmail.com>
On Thu, 5 Feb 2026 08:58:40 -0500
"Seokwoo Chung (Ryan)" <seokwoo.chung130@gmail.com> wrote:
> register_fprobe() passes its filter and notfilter strings directly to
> glob_match(), which only understands shell-style globs (*, ?, [...]).
> Comma-separated symbol lists such as "vfs_read,vfs_open" never match
> any symbol because no kernel symbol contains a comma.
>
> Add glob_match_comma_list() that splits the filter on commas and
> checks each entry individually with glob_match(). The existing
> single-pattern fast path is preserved (no commas means the loop
> executes exactly once).
>
> This is required by the comma-separated fprobe list syntax introduced
> in the preceding patch; without it, enabling a list-mode fprobe event
> fails with "Could not enable event".
OK, in this case, you should reorder patch this as the first one.
Please make this [1/4] and remove this requirement explanation paragraph.
The patch itself looks good to me.
Thank you,
>
> Signed-off-by: Seokwoo Chung (Ryan) <seokwoo.chung130@gmail.com>
> ---
> kernel/trace/fprobe.c | 30 ++++++++++++++++++++++++++++--
> 1 file changed, 28 insertions(+), 2 deletions(-)
>
> diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c
> index 1188eefef07c..2acd24b80d04 100644
> --- a/kernel/trace/fprobe.c
> +++ b/kernel/trace/fprobe.c
> @@ -672,12 +672,38 @@ struct filter_match_data {
> struct module **mods;
> };
>
> +/*
> + * Check if @name matches any comma-separated glob pattern in @list.
> + * If @list contains no commas, this is equivalent to glob_match().
> + */
> +static bool glob_match_comma_list(const char *list, const char *name)
> +{
> + const char *cur = list;
> +
> + while (*cur) {
> + const char *sep = strchr(cur, ',');
> + int len = sep ? sep - cur : strlen(cur);
> + char pat[KSYM_NAME_LEN];
> +
> + if (len > 0 && len < KSYM_NAME_LEN) {
> + memcpy(pat, cur, len);
> + pat[len] = '\0';
> + if (glob_match(pat, name))
> + return true;
> + }
> + if (!sep)
> + break;
> + cur = sep + 1;
> + }
> + return false;
> +}
> +
> static int filter_match_callback(void *data, const char *name, unsigned long addr)
> {
> struct filter_match_data *match = data;
>
> - if (!glob_match(match->filter, name) ||
> - (match->notfilter && glob_match(match->notfilter, name)))
> + if (!glob_match_comma_list(match->filter, name) ||
> + (match->notfilter && glob_match_comma_list(match->notfilter, name)))
> return 0;
>
> if (!ftrace_location(addr))
> --
> 2.43.0
>
>
--
Masami Hiramatsu (Google) <mhiramat@kernel.org>
next prev parent reply other threads:[~2026-03-24 1:59 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-05 13:58 [PATCH v6 0/4] tracing/fprobe: Support comma-separated symbol lists and :entry/:exit suffixes Seokwoo Chung (Ryan)
2026-02-05 13:58 ` [PATCH v6 1/4] tracing/fprobe: Support comma-separated symbols and :entry/:exit Seokwoo Chung (Ryan)
2026-03-24 1:50 ` Masami Hiramatsu
2026-02-05 13:58 ` [PATCH v6 2/4] fprobe: Support comma-separated filters in register_fprobe() Seokwoo Chung (Ryan)
2026-03-24 1:59 ` Masami Hiramatsu [this message]
2026-02-05 13:58 ` [PATCH v6 3/4] docs: tracing/fprobe: Document list filters and :entry/:exit Seokwoo Chung (Ryan)
2026-03-24 4:07 ` Masami Hiramatsu
2026-02-05 13:58 ` [PATCH v6 4/4] selftests/ftrace: Add accept cases for fprobe list syntax Seokwoo Chung (Ryan)
2026-03-24 4:12 ` Masami Hiramatsu
2026-03-24 1:51 ` [PATCH v6 0/4] tracing/fprobe: Support comma-separated symbol lists and :entry/:exit suffixes Masami Hiramatsu
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=20260324105913.cf60a2f2fcc64afbe30bd22e@kernel.org \
--to=mhiramat@kernel.org \
--cc=corbet@lwn.net \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=rostedt@goodmis.org \
--cc=seokwoo.chung130@gmail.com \
--cc=shuah@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