Hi Sakari, Laurent,
On Tue, 2021-01-05 at 02:20 +0200, Sakari Ailus wrote:
> Hi Ezequiel,
>
> On Mon, Jan 04, 2021 at 02:48:40PM -0300, Ezequiel Garcia wrote:
> > There is currently little to none information available
> > about the reasons why a v4l2-async device hasn't
> > probed completely.
> >
> > Inspired by the "devices_deferred" debugfs file,
> > add a file to list information about the subdevices
> > that are on waiting lists, for each notifier.
> >
> > This is useful to debug v4l2-async subdevices
> > and notifiers, for instance when doing device bring-up.
> >
> > For instance, a typical output would be:
> >
> > $ cat /sys/kernel/debug/video4linux/pending_async_subdevices
> > ipu1_csi1:
> > [fwnode] dev=20e0000.iomuxc-gpr:ipu1_csi1_mux, node=/soc/bus@2000000/iomuxc-gpr@20e0000/ipu1_csi1_mux
> > ipu1_csi0:
> > [fwnode] dev=20e0000.iomuxc-gpr:ipu1_csi0_mux, node=/soc/bus@2000000/iomuxc-gpr@20e0000/ipu1_csi0_mux
> > imx6-mipi-csi2:
> > [fwnode] dev=1-003c, node=/soc/bus@2100000/i2c@21a4000/camera@3c
> > imx-media:
> >
> > Note that match-type "custom" prints no information.
> > Since there are no in-tree users of this match-type,
> > the implementation doesn't bother.
> >
> > Signed-off-by: Ezequiel Garcia <ezequiel@xxxxxxxxxxxxx>
> > ---
> > v2:
> > * Print fwnode path, as suggested by Sakari.
> > * Print the subdevices under their corresponding notifier.
> > * Rename the file as suggested by Laurent.
> > ---
> > drivers/media/v4l2-core/v4l2-async.c | 66 ++++++++++++++++++++++++++++
> > drivers/media/v4l2-core/v4l2-dev.c | 5 +++
> > include/media/v4l2-async.h | 9 ++++
> > 3 files changed, 80 insertions(+)
> >
> > diff --git a/drivers/media/v4l2-core/v4l2-async.c b/drivers/media/v4l2-core/v4l2-async.c
> > index e3ab003a6c85..d779808abb3b 100644
> > --- a/drivers/media/v4l2-core/v4l2-async.c
> > +++ b/drivers/media/v4l2-core/v4l2-async.c
> > @@ -5,6 +5,7 @@
> > * Copyright (C) 2012-2013, Guennadi Liakhovetski <g.liakhovetski@xxxxxx>
> > */
> >
> > +#include <linux/debugfs.h>
> > #include <linux/device.h>
> > #include <linux/err.h>
> > #include <linux/i2c.h>
> > @@ -14,6 +15,7 @@
> > #include <linux/mutex.h>
> > #include <linux/of.h>
> > #include <linux/platform_device.h>
> > +#include <linux/seq_file.h>
> > #include <linux/slab.h>
> > #include <linux/types.h>
> >
> > @@ -837,3 +839,67 @@ void v4l2_async_unregister_subdev(struct v4l2_subdev *sd)
> > mutex_unlock(&list_lock);
> > }
> > EXPORT_SYMBOL(v4l2_async_unregister_subdev);
> > +
> > +static void print_waiting_subdev(struct seq_file *s,
> > + struct v4l2_async_subdev *asd)
> > +{
> > + switch (asd->match_type) {
> > + case V4L2_ASYNC_MATCH_CUSTOM:
> > + seq_puts(s, " [custom]\n");
> > + break;
> > + case V4L2_ASYNC_MATCH_DEVNAME:
> > + seq_printf(s, " [devname] %s\n", asd->match.device_name);
> > + break;
> > + case V4L2_ASYNC_MATCH_I2C:
> > + seq_printf(s, " [i2c] %d-%04x\n", asd->match.i2c.adapter_id,
> > + asd->match.i2c.address);
> > + break;
> > + case V4L2_ASYNC_MATCH_FWNODE: {
> > + struct fwnode_handle *fwnode = asd->match.fwnode;
> > +
> > + if (fwnode_graph_is_endpoint(fwnode))
> > + fwnode = fwnode_graph_get_port_parent(fwnode);
>
> Could you print the endpoint node name as-is? That's what matching uses,
> too. You'd need one more local variable for that I think.
>
Since we are printing the full path for the node, how about this:
devnode = fwnode_graph_is_endpoint(fwnode) ?
fwnode_graph_get_port_parent(fwnode) : fwnode;
seq_printf(s, " [fwnode] dev=%s, node=%pfw\n",
devnode->dev ? dev_name(devnode->dev) : "nil",
fwnode);
The parent is used only to print a more useful dev_name,
but otherwise the actual full path is used,
whether it's an endpoint or not.
Ezequiel
