On Mon Jun 23 13:00:33 2025 +0200, Hans de Goede wrote:
> The ov7251 driver waits for the endpoint fwnode to show up in case this
> fwnode is created by a bridge-driver.
>
> It does this by returning -EPROBE_DEFER, but it does not use
> dev_err_probe() so no reason for deferring gets registered.
>
> After 30 seconds the kernel logs a warning that the probe is still
> deferred, which looks like this:
>
> [ 33.952052] i2c i2c-INT347E:00: deferred probe pending: (reason unknown)
>
> Use dev_err_probe() when returning -EPROBE_DEFER to register the probe
> deferral reason changing the error to:
>
> deferred probe pending: waiting for fwnode graph endpoint
>
> Also update the comment to not refer to the no longer existing cio2-bridge
> code.
>
> Signed-off-by: Hans de Goede <hdego...@redhat.com>
> Signed-off-by: Sakari Ailus <sakari.ai...@linux.intel.com>
> Signed-off-by: Hans Verkuil <hverk...@xs4all.nl>
Patch committed.
Thanks,
Hans Verkuil
drivers/media/i2c/ov7251.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
---
diff --git a/drivers/media/i2c/ov7251.c b/drivers/media/i2c/ov7251.c
index 3226888d77e9..31a42d81e970 100644
--- a/drivers/media/i2c/ov7251.c
+++ b/drivers/media/i2c/ov7251.c
@@ -1486,9 +1486,14 @@ static int ov7251_check_hwcfg(struct ov7251 *ov7251)
unsigned int i, j;
int ret;
+ /*
+ * Sometimes the fwnode graph is initialized by the bridge driver
+ * Bridge drivers doing this may also add GPIO mappings, wait for this.
+ */
endpoint = fwnode_graph_get_next_endpoint(fwnode, NULL);
if (!endpoint)
- return -EPROBE_DEFER; /* could be provided by cio2-bridge */
+ return dev_err_probe(ov7251->dev, -EPROBE_DEFER,
+ "waiting for fwnode graph endpoint\n");
ret = v4l2_fwnode_endpoint_alloc_parse(endpoint, &bus_cfg);
fwnode_handle_put(endpoint);
