From: Daniel Scally <djrscally@gmail.com>
To: linux-media@vger.kernel.org
Cc: yong.zhi@intel.com, sakari.ailus@linux.intel.com,
bingbu.cao@intel.com, tian.shu.qiu@intel.com
Subject: [PATCH] media: ipu3-cio2: Move functionality from .complete() to .bound()
Date: Fri, 6 May 2022 22:15:55 +0100 [thread overview]
Message-ID: <20220506211555.1364864-1-djrscally@gmail.com> (raw)
Creating links and registering subdev nodes during the .complete()
callback has the unfortunate effect of preventing all cameras that
connect to a notifier from working if any one of their drivers fails
to probe. Moving the functionality from .complete() to .bound() allows
those camera sensor drivers that did probe correctly to work regardless.
Signed-off-by: Daniel Scally <djrscally@gmail.com>
---
This results in v4l2_device_register_subdev_nodes() being called multiple times
but since it's a no-op where the devnode exists already, I think that it's ok.
drivers/media/pci/intel/ipu3/ipu3-cio2-main.c | 65 +++++++------------
1 file changed, 23 insertions(+), 42 deletions(-)
diff --git a/drivers/media/pci/intel/ipu3/ipu3-cio2-main.c b/drivers/media/pci/intel/ipu3/ipu3-cio2-main.c
index 93cc0577b6db..6a1bcb20725d 100644
--- a/drivers/media/pci/intel/ipu3/ipu3-cio2-main.c
+++ b/drivers/media/pci/intel/ipu3/ipu3-cio2-main.c
@@ -1387,7 +1387,10 @@ static int cio2_notifier_bound(struct v4l2_async_notifier *notifier,
{
struct cio2_device *cio2 = to_cio2_device(notifier);
struct sensor_async_subdev *s_asd = to_sensor_asd(asd);
+ struct device *dev = &cio2->pci_dev->dev;
struct cio2_queue *q;
+ unsigned int pad;
+ int ret;
if (cio2->queue[s_asd->csi2.port].sensor)
return -EBUSY;
@@ -1398,7 +1401,26 @@ static int cio2_notifier_bound(struct v4l2_async_notifier *notifier,
q->sensor = sd;
q->csi_rx_base = cio2->base + CIO2_REG_PIPE_BASE(q->csi2.port);
- return 0;
+ for (pad = 0; pad < q->sensor->entity.num_pads; pad++)
+ if (q->sensor->entity.pads[pad].flags &
+ MEDIA_PAD_FL_SOURCE)
+ break;
+
+ if (pad == q->sensor->entity.num_pads) {
+ dev_err(dev, "failed to find src pad for %s\n",
+ q->sensor->name);
+ return -ENXIO;
+ }
+
+ ret = media_create_pad_link(&q->sensor->entity, pad, &q->subdev.entity,
+ CIO2_PAD_SINK, 0);
+ if (ret) {
+ dev_err(dev, "failed to create link for %s\n",
+ q->sensor->name);
+ return ret;
+ }
+
+ return v4l2_device_register_subdev_nodes(&cio2->v4l2_dev);
}
/* The .unbind callback */
@@ -1412,50 +1434,9 @@ static void cio2_notifier_unbind(struct v4l2_async_notifier *notifier,
cio2->queue[s_asd->csi2.port].sensor = NULL;
}
-/* .complete() is called after all subdevices have been located */
-static int cio2_notifier_complete(struct v4l2_async_notifier *notifier)
-{
- struct cio2_device *cio2 = to_cio2_device(notifier);
- struct device *dev = &cio2->pci_dev->dev;
- struct sensor_async_subdev *s_asd;
- struct v4l2_async_subdev *asd;
- struct cio2_queue *q;
- unsigned int pad;
- int ret;
-
- list_for_each_entry(asd, &cio2->notifier.asd_list, asd_list) {
- s_asd = to_sensor_asd(asd);
- q = &cio2->queue[s_asd->csi2.port];
-
- for (pad = 0; pad < q->sensor->entity.num_pads; pad++)
- if (q->sensor->entity.pads[pad].flags &
- MEDIA_PAD_FL_SOURCE)
- break;
-
- if (pad == q->sensor->entity.num_pads) {
- dev_err(dev, "failed to find src pad for %s\n",
- q->sensor->name);
- return -ENXIO;
- }
-
- ret = media_create_pad_link(
- &q->sensor->entity, pad,
- &q->subdev.entity, CIO2_PAD_SINK,
- 0);
- if (ret) {
- dev_err(dev, "failed to create link for %s\n",
- q->sensor->name);
- return ret;
- }
- }
-
- return v4l2_device_register_subdev_nodes(&cio2->v4l2_dev);
-}
-
static const struct v4l2_async_notifier_operations cio2_async_ops = {
.bound = cio2_notifier_bound,
.unbind = cio2_notifier_unbind,
- .complete = cio2_notifier_complete,
};
static int cio2_parse_firmware(struct cio2_device *cio2)
--
2.25.1
next reply other threads:[~2022-05-06 21:16 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-06 21:15 Daniel Scally [this message]
2022-06-06 17:04 ` [PATCH] media: ipu3-cio2: Move functionality from .complete() to .bound() Daniel Scally
2022-06-07 7:35 ` Kieran Bingham
2022-06-07 8:47 ` Daniel Scally
2022-06-07 7:05 ` Jacopo Mondi
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=20220506211555.1364864-1-djrscally@gmail.com \
--to=djrscally@gmail.com \
--cc=bingbu.cao@intel.com \
--cc=linux-media@vger.kernel.org \
--cc=sakari.ailus@linux.intel.com \
--cc=tian.shu.qiu@intel.com \
--cc=yong.zhi@intel.com \
/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