[PATCH] platform/chrome: cros_usbpd_logger: fail probe if HC not supported

[PATCH] platform/chrome: cros_usbpd_logger: fail probe if HC not supported

[Daniel Schaefer]

If the EC_CMD_PD_GET_LOG_ENTRY does not exist, the driver does not need
to stay active.

Framework Computer systems don't use the PD infrastructure in the chrome
EC firmware, so it does not support this host command.

Previously the driver probe would always succeed and periodically keep
trying to use this host command, resulting in unnecessary EC traffic and
EC log error messages.

Cc: Benson Leung <bleung@chromium.org>
Cc: Tzung-Bi Shih <tzungbi@kernel.org>
Cc: Dustin L. Howett <dustin@howett.net>
Cc: <linux@frame.work>
Signed-off-by: Daniel Schaefer <dhs@frame.work>
---

With this patch the following dmesg log appears on Framework systems.
[   25.171086] cros-usbpd-logger cros-usbpd-logger.8.auto: host command not supported.

 drivers/platform/chrome/cros_usbpd_logger.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/platform/chrome/cros_usbpd_logger.c b/drivers/platform/chrome/cros_usbpd_logger.c
index 7ce75e2e039e..df6b1d97589a 100644
--- a/drivers/platform/chrome/cros_usbpd_logger.c
+++ b/drivers/platform/chrome/cros_usbpd_logger.c
@@ -207,6 +207,11 @@ static int cros_usbpd_logger_probe(struct platform_device *pd)
 	logger->dev = dev;
 	logger->ec_dev = ec_dev;
 
+	if (-EOPNOTSUPP == PTR_ERR(ec_get_log_entry(logger))) {
+		dev_info(dev, "host command not supported.\n");
+		return -ENODEV;
+	}
+
 	platform_set_drvdata(pd, logger);
 
 	/* Retrieve PD event logs periodically */
-- 
2.34.1

Re: [PATCH] platform/chrome: cros_usbpd_logger: fail probe if HC not supported

[Tzung-Bi Shih]

On Thu, Feb 20, 2025 at 10:34:41AM +0800, Daniel Schaefer wrote:
> If the EC_CMD_PD_GET_LOG_ENTRY does not exist, the driver does not need
> to stay active.
> 
> Framework Computer systems don't use the PD infrastructure in the chrome
> EC firmware, so it does not support this host command.

Does the firmware still report `EC_FEATURE_USB_PD` feature [1]?  Similarly,
do "cros-charge-control" and "cros-usbpd-charger" need to stay active [2]?

[1]: https://elixir.bootlin.com/linux/v6.13/source/drivers/mfd/cros_ec_dev.c#L263
[2]: https://elixir.bootlin.com/linux/v6.13/source/drivers/mfd/cros_ec_dev.c#L89

Re: [PATCH] platform/chrome: cros_usbpd_logger: fail probe if HC not supported

[Daniel Schaefer]

You are right, if we remove this feature, the driver won't load. So we
don't need this kernel patch.
We found there's a bug in Chrome EC where it enables this feature,
even when the Kconfig is turned off.
We're including this patch in our future Framework EC builds:
https://github.com/FrameworkComputer/EmbeddedController/commit/57030a466bb7e130313cf2b2a305a753e9c9a55e

On Thu, Feb 20, 2025 at 5:48 PM Tzung-Bi Shih <tzungbi@kernel.org> wrote:
>
> On Thu, Feb 20, 2025 at 10:34:41AM +0800, Daniel Schaefer wrote:
> > If the EC_CMD_PD_GET_LOG_ENTRY does not exist, the driver does not need
> > to stay active.
> >
> > Framework Computer systems don't use the PD infrastructure in the chrome
> > EC firmware, so it does not support this host command.
>
> Does the firmware still report `EC_FEATURE_USB_PD` feature [1]?  Similarly,
> do "cros-charge-control" and "cros-usbpd-charger" need to stay active [2]?
>
> [1]: https://elixir.bootlin.com/linux/v6.13/source/drivers/mfd/cros_ec_dev.c#L263
> [2]: https://elixir.bootlin.com/linux/v6.13/source/drivers/mfd/cros_ec_dev.c#L89