linux-kernel - [PATCH v2] media: mipi-csis: Emit V4L2_EVENT_FRAME

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: <20240314093652.56923-1-stefan.klug@ideasonboard.com>
Date: Thu, 14 Mar 2024 10:36:50 +0100
From: Stefan Klug <stefan.klug@...asonboard.com>
To: libcamera-devel@...ts.libcamera.org,
	linux-media@...r.kernel.org,
	linux-arm-kernel@...ts.infradead.org,
	linux-kernel@...r.kernel.org
Cc: Stefan Klug <stefan.klug@...asonboard.com>,
	Laurent Pinchart <laurent.pinchart@...asonboard.com>,
	Jacopo Mondi <jacopo.mondi@...asonboard.com>,
	Umang Jain <umang.jain@...asonboard.com>,
	Rui Miguel Silva <rmfrfs@...il.com>,
	Martin Kepplinger <martink@...teo.de>,
	Purism Kernel Team <kernel@...i.sm>,
	Mauro Carvalho Chehab <mchehab@...nel.org>,
	Shawn Guo <shawnguo@...nel.org>,
	Sascha Hauer <s.hauer@...gutronix.de>,
	Pengutronix Kernel Team <kernel@...gutronix.de>,
	Fabio Estevam <festevam@...il.com>,
	NXP Linux Team <linux-imx@....com>
Subject: [PATCH v2] media: mipi-csis: Emit V4L2_EVENT_FRAME_SYNC events

The Samsung CSIS MIPI receiver provides a start-of-frame interrupt and
a framecount register. As the CSI receiver is the hardware unit that lies
closest to the sensor, the frame counter is the best we can get on these
devices. In case of the ISI available on the i.MX8 M Plus it is also the
only native start-of-frame signal available.
This patch exposes the sof interrupt and the framecount as
V4L2_EVENT_FRAME_SYNC event on the subdevice.

It was tested on a Debix-Som-A with a 6.8-rc4 kernel.

Reviewed-by: Laurent Pinchart <laurent.pinchart@...asonboard.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@...asonboard.com>
Reviewed-by: Umang Jain <umang.jain@...asonboard.com>
Signed-off-by: Stefan Klug <stefan.klug@...asonboard.com>
---
Changes v1 -> v2:
 - fixed formatting issues from review
 - moved frame variable declaration to top of subscribe_event()

Thanks all for the review!

 drivers/media/platform/nxp/imx-mipi-csis.c | 34 +++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/drivers/media/platform/nxp/imx-mipi-csis.c b/drivers/media/platform/nxp/imx-mipi-csis.c
index db8ff5f5c4d3..664be27c4224 100644
--- a/drivers/media/platform/nxp/imx-mipi-csis.c
+++ b/drivers/media/platform/nxp/imx-mipi-csis.c
@@ -30,6 +30,7 @@
 
 #include <media/v4l2-common.h>
 #include <media/v4l2-device.h>
+#include <media/v4l2-event.h>
 #include <media/v4l2-fwnode.h>
 #include <media/v4l2-mc.h>
 #include <media/v4l2-subdev.h>
@@ -742,6 +743,18 @@ static void mipi_csis_stop_stream(struct mipi_csis_device *csis)
 	mipi_csis_system_enable(csis, false);
 }
 
+static void mipi_csis_queue_event_sof(struct mipi_csis_device *csis)
+{
+	u32 frame;
+	struct v4l2_event event = {
+		.type = V4L2_EVENT_FRAME_SYNC,
+	};
+
+	frame = mipi_csis_read(csis, MIPI_CSIS_FRAME_COUNTER_CH(0));
+	event.u.frame_sync.frame_sequence = frame;
+	v4l2_event_queue(csis->sd.devnode, &event);
+}
+
 static irqreturn_t mipi_csis_irq_handler(int irq, void *dev_id)
 {
 	struct mipi_csis_device *csis = dev_id;
@@ -765,6 +778,10 @@ static irqreturn_t mipi_csis_irq_handler(int irq, void *dev_id)
 				event->counter++;
 		}
 	}
+
+	if (status & MIPI_CSIS_INT_SRC_FRAME_START)
+		mipi_csis_queue_event_sof(csis);
+
 	spin_unlock_irqrestore(&csis->slock, flags);
 
 	mipi_csis_write(csis, MIPI_CSIS_INT_SRC, status);
@@ -1154,8 +1171,23 @@ static int mipi_csis_log_status(struct v4l2_subdev *sd)
 	return 0;
 }
 
+static int mipi_csis_subscribe_event(struct v4l2_subdev *sd, struct v4l2_fh *fh,
+				     struct v4l2_event_subscription *sub)
+{
+	if (sub->type != V4L2_EVENT_FRAME_SYNC)
+		return -EINVAL;
+
+	/* V4L2_EVENT_FRAME_SYNC doesn't require an id, so zero should be set */
+	if (sub->id != 0)
+		return -EINVAL;
+
+	return v4l2_event_subscribe(fh, sub, 0, NULL);
+}
+
 static const struct v4l2_subdev_core_ops mipi_csis_core_ops = {
 	.log_status	= mipi_csis_log_status,
+	.subscribe_event =  mipi_csis_subscribe_event,
+	.unsubscribe_event = v4l2_event_subdev_unsubscribe,
 };
 
 static const struct v4l2_subdev_video_ops mipi_csis_video_ops = {
@@ -1358,7 +1390,7 @@ static int mipi_csis_subdev_init(struct mipi_csis_device *csis)
 	snprintf(sd->name, sizeof(sd->name), "csis-%s",
 		 dev_name(csis->dev));
 
-	sd->flags |= V4L2_SUBDEV_FL_HAS_DEVNODE;
+	sd->flags |= V4L2_SUBDEV_FL_HAS_DEVNODE | V4L2_SUBDEV_FL_HAS_EVENTS;
 	sd->ctrl_handler = NULL;
 
 	sd->entity.function = MEDIA_ENT_F_VID_IF_BRIDGE;
-- 
2.40.1