From: "Masami Hiramatsu (Google)" <mhiramat@kernel.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org
Subject: [PATCH 0/4] tracing: fprobe-events: Register fprobe only when the event is enabled
Date: Sun, 16 Mar 2025 21:21:11 +0900 [thread overview]
Message-ID: <174212767109.348872.18231451508464729427.stgit@devnote2> (raw)
Hi,
Here is a series of patches to register fprobe only when the fprobe
event is enabled. Steve suggested the fprobe-events are always
registered when it is defined, even if it is disabled and that makes
system overhead. This series registeres the fprobes only when the event
is enabled.
NOTE: tracepoint has a AB-BA locking issue[*], so it is still enabled
when the event is defined. That should be eventually solved, but it may
need more complicated change. Thus this series focuses on registering
fprobe when the event is enabled.
(*) AB-BA lock if we enable tracepoint when enabling tprobe event:
[event enable]
__ftrace_event_enable_disable() ----> event_mutex
__regsiter_tracepoint_fprobe()
find_tracepoint()
for_each_module_tracepoint() ----> tracepoint_module_list_mutex
[module loading]
prepare_coming_module()
tracepoint_module_notify()
tracepoint_module_coming() ----> tracepoint_module_list_mutex
__tracepoint_probe_module_cb() ---> event_mutex
I have an idea to defer fprobe event enablement when module loading
so that we can avoid taking event_mutex in
__tracepoint_probe_module_cb(), but it needs to introduce another
list of tracepoint_user and another mutex.
Thank you,
---
Masami Hiramatsu (Google) (4):
tracing: tprobe-events: Remove mod field from tprobe-event
tracing: tprobe-events: Support multiple tprobes on the same tracepoint
tracing: fprobe-events: Register fprobe-events only when it is enabled
selftests: tracing: Enable fprobe events before checking enable_functions
include/linux/fprobe.h | 8
include/linux/module.h | 4
kernel/trace/fprobe.c | 29 +
kernel/trace/trace_fprobe.c | 490 +++++++++++++-------
.../ftrace/test.d/dynevent/add_remove_fprobe.tc | 30 +
5 files changed, 362 insertions(+), 199 deletions(-)
--
Masami Hiramatsu (Google) <mhiramat@kernel.org>
next reply other threads:[~2025-03-16 12:21 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-16 12:21 Masami Hiramatsu (Google) [this message]
2025-03-16 12:21 ` [PATCH 1/4] tracing: tprobe-events: Remove mod field from tprobe-event Masami Hiramatsu (Google)
2025-03-16 12:21 ` [PATCH 2/4] tracing: tprobe-events: Support multiple tprobes on the same tracepoint Masami Hiramatsu (Google)
2025-03-25 17:06 ` Steven Rostedt
2025-03-25 22:08 ` Masami Hiramatsu
2025-03-29 13:35 ` Masami Hiramatsu
2025-03-16 12:21 ` [PATCH 3/4] tracing: fprobe-events: Register fprobe-events only when it is enabled Masami Hiramatsu (Google)
2025-03-25 18:41 ` Steven Rostedt
2025-03-25 21:56 ` Masami Hiramatsu
2025-03-16 12:21 ` [PATCH 4/4] selftests: tracing: Enable fprobe events before checking enable_functions Masami Hiramatsu (Google)
2025-03-25 18:42 ` Steven Rostedt
2025-03-25 21:47 ` Masami Hiramatsu
2025-03-17 8:03 ` [RFC PATCH] tracing: tprobe-events: Register tracepoint when enable tprobe event Masami Hiramatsu (Google)
2025-03-17 8:10 ` Masami Hiramatsu
2025-05-01 15:27 ` Steven Rostedt
2025-05-09 23:41 ` 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=174212767109.348872.18231451508464729427.stgit@devnote2 \
--to=mhiramat@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=rostedt@goodmis.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