linux-kernel - [PATCH v2 1/2] perf/ring_buffer: Trigger IO signals for watermark wakeups

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20240305060844.9499-1-khuey@kylehuey.com>
Date: Mon,  4 Mar 2024 22:08:42 -0800
From: Kyle Huey <me@...ehuey.com>
To: Kyle Huey <khuey@...ehuey.com>,
	Robert O'Callahan <robert@...llahan.org>,
	linux-kernel@...r.kernel.org,
	Ian Rodgers <irogers@...gle.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Namhyung Kim <namhyung@...nel.org>
Cc: Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...hat.com>,
	Mark Rutland <mark.rutland@....com>,
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
	Jiri Olsa <jolsa@...nel.org>,
	Adrian Hunter <adrian.hunter@...el.com>,
	linux-perf-users@...r.kernel.org
Subject: [PATCH v2 1/2] perf/ring_buffer: Trigger IO signals for watermark wakeups

perf_output_wakeup() already marks the perf event fd available for polling.
Add code to trigger IO signals with FASYNC too.

Signed-off-by: Kyle Huey <khuey@...ehuey.com>
---
 include/linux/perf_event.h  | 8 ++++++++
 kernel/events/core.c        | 8 --------
 kernel/events/ring_buffer.c | 3 +++
 3 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index c8bd5bb6610c..c077968d7e52 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -1914,4 +1914,12 @@ static inline void perf_lopwr_cb(bool mode)
 }
 #endif
 
+static inline struct fasync_struct **perf_event_fasync(struct perf_event *event)
+{
+	/* only the parent has fasync state */
+	if (event->parent)
+		event = event->parent;
+	return &event->fasync;
+}
+
 #endif /* _LINUX_PERF_EVENT_H */
diff --git a/kernel/events/core.c b/kernel/events/core.c
index a329bec42c4d..36b5fbdf8e6e 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -6684,14 +6684,6 @@ static const struct file_operations perf_fops = {
  * to user-space before waking everybody up.
  */
 
-static inline struct fasync_struct **perf_event_fasync(struct perf_event *event)
-{
-	/* only the parent has fasync state */
-	if (event->parent)
-		event = event->parent;
-	return &event->fasync;
-}
-
 void perf_event_wakeup(struct perf_event *event)
 {
 	ring_buffer_wakeup(event);
diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
index 60ed43d1c29e..033e54bb5c62 100644
--- a/kernel/events/ring_buffer.c
+++ b/kernel/events/ring_buffer.c
@@ -22,6 +22,9 @@ static void perf_output_wakeup(struct perf_output_handle *handle)
 	atomic_set(&handle->rb->poll, EPOLLIN);
 
 	handle->event->pending_wakeup = 1;
+	if (*perf_event_fasync(handle->event) && !handle->event->pending_kill)
+		handle->event->pending_kill = POLL_IN;
+
 	irq_work_queue(&handle->event->pending_irq);
 }
 
-- 
2.34.1