linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@kernel.org>
To: linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org,
	bpf@vger.kernel.org, x86@kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Josh Poimboeuf <jpoimboe@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>, Jiri Olsa <jolsa@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Andrii Nakryiko <andrii@kernel.org>,
	Indu Bhagat <indu.bhagat@oracle.com>,
	"Jose E. Marchesi" <jemarch@gnu.org>,
	Beau Belgrave <beaub@linux.microsoft.com>,
	Jens Remus <jremus@linux.ibm.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Jens Axboe <axboe@kernel.dk>, Florian Weimer <fweimer@redhat.com>,
	Sam James <sam@gentoo.org>
Subject: [PATCH v13 11/14] unwind: Add USED bit to only have one conditional on way back to user space
Date: Mon, 07 Jul 2025 21:22:50 -0400	[thread overview]
Message-ID: <20250708012359.516909531@kernel.org> (raw)
In-Reply-To: 20250708012239.268642741@kernel.org

From: Steven Rostedt <rostedt@goodmis.org>

On the way back to user space, the function unwind_reset_info() is called
unconditionally (but always inlined). It currently has two conditionals.
One that checks the unwind_mask which is set whenever a deferred trace is
called and is used to know that the mask needs to be cleared. The other
checks if the cache has been allocated, and if so, it resets the
nr_entries so that the unwinder knows it needs to do the work to get a new
user space stack trace again (it only does it once per entering the
kernel).

Use one of the bits in the unwind mask as a "USED" bit that gets set
whenever a trace is created. This will make it possible to only check the
unwind_mask in the unwind_reset_info() to know if it needs to do work or
not and eliminates a conditional that happens every time the task goes
back to user space.

Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
---
 include/linux/unwind_deferred.h | 14 +++++++-------
 kernel/unwind/deferred.c        |  5 ++++-
 2 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/include/linux/unwind_deferred.h b/include/linux/unwind_deferred.h
index 587e120c0fd6..376bfd50ff75 100644
--- a/include/linux/unwind_deferred.h
+++ b/include/linux/unwind_deferred.h
@@ -21,6 +21,10 @@ struct unwind_work {
 #define UNWIND_PENDING_BIT	(BITS_PER_LONG - 1)
 #define UNWIND_PENDING		BIT(UNWIND_PENDING_BIT)
 
+/* Set if the unwinding was used (directly or deferred) */
+#define UNWIND_USED_BIT		(UNWIND_PENDING_BIT - 1)
+#define UNWIND_USED		BIT(UNWIND_USED_BIT)
+
 enum {
 	UNWIND_ALREADY_PENDING	= 1,
 	UNWIND_ALREADY_EXECUTED	= 2,
@@ -49,14 +53,10 @@ static __always_inline void unwind_reset_info(void)
 				return;
 		} while (!try_cmpxchg(&info->unwind_mask, &bits, 0UL));
 		current->unwind_info.id.id = 0;
+
+		if (unlikely(info->cache))
+			info->cache->nr_entries = 0;
 	}
-	/*
-	 * As unwind_user_faultable() can be called directly and
-	 * depends on nr_entries being cleared on exit to user,
-	 * this needs to be a separate conditional.
-	 */
-	if (unlikely(info->cache))
-		info->cache->nr_entries = 0;
 }
 
 #else /* !CONFIG_UNWIND_USER */
diff --git a/kernel/unwind/deferred.c b/kernel/unwind/deferred.c
index 256458f3eafe..9299974b6562 100644
--- a/kernel/unwind/deferred.c
+++ b/kernel/unwind/deferred.c
@@ -140,6 +140,9 @@ int unwind_user_faultable(struct unwind_stacktrace *trace)
 
 	cache->nr_entries = trace->nr;
 
+	/* Clear nr_entries on way back to user space */
+	set_bit(UNWIND_USED_BIT, &info->unwind_mask);
+
 	return 0;
 }
 
@@ -314,7 +317,7 @@ int unwind_deferred_init(struct unwind_work *work, unwind_callback_t func)
 	guard(mutex)(&callback_mutex);
 
 	/* See if there's a bit in the mask available */
-	if (unwind_mask == ~(UNWIND_PENDING))
+	if (unwind_mask == ~(UNWIND_PENDING|UNWIND_USED))
 		return -EBUSY;
 
 	work->bit = ffz(unwind_mask);
-- 
2.47.2



  parent reply	other threads:[~2025-07-08  1:23 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-08  1:22 [PATCH v13 00/14] unwind_user: x86: Deferred unwinding infrastructure Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 01/14] unwind_user: Add user space unwinding API Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 02/14] unwind_user: Add frame pointer support Steven Rostedt
2025-07-09 10:01   ` Jens Remus
2025-07-10 12:28     ` Jens Remus
2025-07-10 15:21     ` Steven Rostedt
2025-07-10 15:41       ` Jens Remus
2025-07-10 17:08         ` Steven Rostedt
2025-07-14 12:52           ` Jens Remus
2025-07-08  1:22 ` [PATCH v13 03/14] unwind_user: Add compat mode " Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 04/14] unwind_user/deferred: Add unwind_user_faultable() Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 05/14] unwind_user/deferred: Add unwind cache Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 06/14] unwind_user/deferred: Add deferred unwinding interface Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 07/14] unwind_user/deferred: Make unwind deferral requests NMI-safe Steven Rostedt
2025-07-14 13:29   ` Peter Zijlstra
2025-07-14 14:19     ` Steven Rostedt
2025-07-14 15:05       ` Peter Zijlstra
2025-07-14 15:11         ` Steven Rostedt
2025-07-15  9:09           ` Peter Zijlstra
2025-07-15 12:35             ` Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 08/14] unwind deferred: Use bitmask to determine which callbacks to call Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 09/14] unwind deferred: Use SRCU unwind_deferred_task_work() Steven Rostedt
2025-07-14 13:56   ` Peter Zijlstra
2025-07-14 14:21     ` Steven Rostedt
2025-07-14 15:03       ` Peter Zijlstra
2025-07-08  1:22 ` [PATCH v13 10/14] unwind: Clear unwind_mask on exit back to user space Steven Rostedt
2025-07-15 10:29   ` Peter Zijlstra
2025-07-15 12:49     ` Steven Rostedt
2025-07-15 18:06       ` Steven Rostedt
2025-07-15 18:10         ` Steven Rostedt
2025-07-15 18:26         ` Steven Rostedt
2025-07-15 19:04         ` Peter Zijlstra
2025-07-15 19:01       ` Peter Zijlstra
2025-07-15 17:20     ` Steven Rostedt
2025-07-15 19:07       ` Peter Zijlstra
2025-07-15 22:01     ` Steven Rostedt
2025-07-16 18:26     ` Steven Rostedt
2025-07-16 18:33       ` Steven Rostedt
2025-07-16 19:25         ` Steven Rostedt
2025-07-08  1:22 ` Steven Rostedt [this message]
2025-07-08  1:22 ` [PATCH v13 12/14] unwind: Finish up unwind when a task exits Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 13/14] unwind_user/x86: Enable frame pointer unwinding on x86 Steven Rostedt
2025-07-11  8:43   ` David Laight
2025-07-11 16:11     ` Steven Rostedt
2025-07-08  1:22 ` [PATCH v13 14/14] unwind_user/x86: Enable compat mode " Steven Rostedt

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=20250708012359.516909531@kernel.org \
    --to=rostedt@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=andrii@kernel.org \
    --cc=axboe@kernel.dk \
    --cc=beaub@linux.microsoft.com \
    --cc=bpf@vger.kernel.org \
    --cc=fweimer@redhat.com \
    --cc=indu.bhagat@oracle.com \
    --cc=jemarch@gnu.org \
    --cc=jolsa@kernel.org \
    --cc=jpoimboe@kernel.org \
    --cc=jremus@linux.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mhiramat@kernel.org \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=sam@gentoo.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=x86@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;
as well as URLs for NNTP newsgroup(s).