diff mbox series

[RFC,1/2] platform/surface: aggregator: Defer probing when serdev is not ready

Message ID 20240502040255.655957-3-weifeng.liu.z@gmail.com (mailing list archive)
State Superseded, archived
Headers show
Series Defer probing of SAM if serdev device is not ready | expand

Commit Message

Weifeng Liu May 2, 2024, 4:02 a.m. UTC
This is an attempt to alleviate race conditions in the SAM driver where
essential resources like serial device and GPIO pins are not ready at
the time ssam_serial_hub_probe() is called.  Instead of giving up
probing, a better way would be to defer the probing by returning
-EPROBE_DEFER, allowing the kernel try again later.

However, there is no way of identifying all such cases from other real
errors in a few days.  So let's take a gradual approach identify and
address these cases as they arise.  This commit marks the initial step
in this process.

Suggested-by: Maximilian Luz <luzmaximilian@gmail.com>
Signed-off-by: Weifeng Liu <weifeng.liu.z@gmail.com>
---
 drivers/platform/surface/aggregator/core.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

Comments

Andy Shevchenko May 2, 2024, 3:19 p.m. UTC | #1
On Thu, May 02, 2024 at 12:02:46PM +0800, Weifeng Liu wrote:
> This is an attempt to alleviate race conditions in the SAM driver where
> essential resources like serial device and GPIO pins are not ready at
> the time ssam_serial_hub_probe() is called.  Instead of giving up
> probing, a better way would be to defer the probing by returning
> -EPROBE_DEFER, allowing the kernel try again later.
> 
> However, there is no way of identifying all such cases from other real
> errors in a few days.  So let's take a gradual approach identify and
> address these cases as they arise.  This commit marks the initial step
> in this process.

...

> +	/* The following step can fail when it's called too early before the
> +	 * underlying uart device is ready (in this case -ENXIO is returned).

UART

> +	 * Instead of simply giving up and losing everything, we can defer
> +	 * the probing by returning -EPROBE_DEFER so that the kernel would be
> +	 * able to retry later. */

/*
 * Use correct style for the comments.
 * Here is the example.
 */

>  	status = serdev_device_open(serdev);
> -	if (status)
> +	if (status == -ENXIO)
> +		status = -EPROBE_DEFER;
> +	if (status) {
> +		dev_err_probe(&serdev->dev, status,
> +			      "failed to open serdev device\n");
>  		goto err_devopen;
> +	}

...

Hans, not sure if it helps, but we added similar into I²C and SPI code.

2dea645ffc21 ("i2c: acpi: Return error pointers from i2c_acpi_new_device()")

9c22ec4ac27b ("spi: Return deferred probe error when controller isn't yet available")
diff mbox series

Patch

diff --git a/drivers/platform/surface/aggregator/core.c b/drivers/platform/surface/aggregator/core.c
index 9591a28bc38a..72a521dd729c 100644
--- a/drivers/platform/surface/aggregator/core.c
+++ b/drivers/platform/surface/aggregator/core.c
@@ -645,9 +645,20 @@  static int ssam_serial_hub_probe(struct serdev_device *serdev)
 	/* Set up serdev device. */
 	serdev_device_set_drvdata(serdev, ctrl);
 	serdev_device_set_client_ops(serdev, &ssam_serdev_ops);
+
+	/* The following step can fail when it's called too early before the
+	 * underlying uart device is ready (in this case -ENXIO is returned).
+	 * Instead of simply giving up and losing everything, we can defer
+	 * the probing by returning -EPROBE_DEFER so that the kernel would be
+	 * able to retry later. */
 	status = serdev_device_open(serdev);
-	if (status)
+	if (status == -ENXIO)
+		status = -EPROBE_DEFER;
+	if (status) {
+		dev_err_probe(&serdev->dev, status,
+			      "failed to open serdev device\n");
 		goto err_devopen;
+	}
 
 	astatus = ssam_serdev_setup_via_acpi(ssh->handle, serdev);
 	if (ACPI_FAILURE(astatus)) {