Linux Trace Kernel
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: LKML <linux-kernel@vger.kernel.org>,
	Linux trace kernel <linux-trace-kernel@vger.kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Yafang Shao <laoar.shao@gmail.com>
Subject: [PATCH] tracing: Add generic test_recursion_try_acquire()
Date: Fri, 14 Apr 2023 22:17:02 -0400	[thread overview]
Message-ID: <20230414221702.554c44fe@rorschach.local.home> (raw)

From: "Steven Rostedt (Google)" <rostedt@goodmis.org>

The ftrace_test_recursion_trylock() also disables preemption. This is not
required, but was a clean up as every place that called it also disabled
preemption, and making the two tightly coupled appeared to make the code
simpler.

But the recursion protection can be used for other purposes that do not
require disabling preemption. As the recursion bits are attached to the
task_struct, it follows the task, so there's no need for preemption being
disabled.

Add test_recursion_try_acquire/release() functions to be used generically,
and separate it from being associated with ftrace. It also removes the
"lock" name, as there is no lock happening. Keeping the "lock" for the
ftrace version is better, as it at least differentiates that preemption is
being disabled (hence, "locking the CPU").

Link: https://lore.kernel.org/linux-trace-kernel/20230321020103.13494-1-laoar.shao@gmail.com/

Cc: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
---
 include/linux/trace_recursion.h | 47 ++++++++++++++++++++++++---------
 kernel/trace/ftrace.c           |  2 ++
 2 files changed, 37 insertions(+), 12 deletions(-)

diff --git a/include/linux/trace_recursion.h b/include/linux/trace_recursion.h
index d48cd92d2364..80de2ee7b4c3 100644
--- a/include/linux/trace_recursion.h
+++ b/include/linux/trace_recursion.h
@@ -150,9 +150,6 @@ extern void ftrace_record_recursion(unsigned long ip, unsigned long parent_ip);
 # define trace_warn_on_no_rcu(ip)	false
 #endif
 
-/*
- * Preemption is promised to be disabled when return bit >= 0.
- */
 static __always_inline int trace_test_and_set_recursion(unsigned long ip, unsigned long pip,
 							int start)
 {
@@ -182,18 +179,11 @@ static __always_inline int trace_test_and_set_recursion(unsigned long ip, unsign
 	val |= 1 << bit;
 	current->trace_recursion = val;
 	barrier();
-
-	preempt_disable_notrace();
-
 	return bit;
 }
 
-/*
- * Preemption will be enabled (if it was previously enabled).
- */
 static __always_inline void trace_clear_recursion(int bit)
 {
-	preempt_enable_notrace();
 	barrier();
 	trace_recursion_clear(bit);
 }
@@ -205,12 +195,18 @@ static __always_inline void trace_clear_recursion(int bit)
  * tracing recursed in the same context (normal vs interrupt),
  *
  * Returns: -1 if a recursion happened.
- *           >= 0 if no recursion.
+ *           >= 0 if no recursion and preemption will be disabled.
  */
 static __always_inline int ftrace_test_recursion_trylock(unsigned long ip,
 							 unsigned long parent_ip)
 {
-	return trace_test_and_set_recursion(ip, parent_ip, TRACE_FTRACE_START);
+	int bit;
+
+	bit = trace_test_and_set_recursion(ip, parent_ip, TRACE_FTRACE_START);
+	if (unlikely(bit < 0))
+		return bit;
+	preempt_disable_notrace();
+	return bit;
 }
 
 /**
@@ -220,6 +216,33 @@ static __always_inline int ftrace_test_recursion_trylock(unsigned long ip,
  * This is used at the end of a ftrace callback.
  */
 static __always_inline void ftrace_test_recursion_unlock(int bit)
+{
+	preempt_enable_notrace();
+	trace_clear_recursion(bit);
+}
+
+/**
+ * test_recursion_try_acquire - tests for recursion in same context
+ *
+ * This will detect recursion of a function.
+ *
+ * Returns: -1 if a recursion happened.
+ *           >= 0 if no recursion
+ */
+static __always_inline int test_recursion_try_acquire(unsigned long ip,
+						      unsigned long parent_ip)
+{
+	return trace_test_and_set_recursion(ip, parent_ip, TRACE_FTRACE_START);
+}
+
+/**
+ * test_recursion_release - called after a success of test_recursion_try_acquire()
+ * @bit: The return of a successful test_recursion_try_acquire()
+ *
+ * This releases the recursion lock taken by a non-negative return call
+ * by test_recursion_try_acquire().
+ */
+static __always_inline void test_recursion_release(int bit)
 {
 	trace_clear_recursion(bit);
 }
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index db8532a4d5c8..1b117ab057ed 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -7299,6 +7299,7 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip,
 	if (bit < 0)
 		return;
 
+	preempt_disable();
 	do_for_each_ftrace_op(op, ftrace_ops_list) {
 		/* Stub functions don't need to be called nor tested */
 		if (op->flags & FTRACE_OPS_FL_STUB)
@@ -7320,6 +7321,7 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip,
 		}
 	} while_for_each_ftrace_op(op);
 out:
+	preempt_enable();
 	trace_clear_recursion(bit);
 }
 
-- 
2.39.2


             reply	other threads:[~2023-04-15  2:17 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-15  2:17 Steven Rostedt [this message]
2023-04-15 13:33 ` [PATCH] tracing: Add generic test_recursion_try_acquire() Yafang Shao
2023-04-15 14:33   ` Yafang Shao
2023-04-15 15:46     ` Steven Rostedt
2023-04-16  2:42       ` Yafang Shao

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=20230414221702.554c44fe@rorschach.local.home \
    --to=rostedt@goodmis.org \
    --cc=laoar.shao@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mhiramat@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