If the chip indicates its "Connection Orientation" standard output control
in STANDARD_OUTPUT_CAPABILITIES register, it can do the thing by
programming CONFIG_STANDARD_OUTPUT register. Due to the optional feature,
the chip which not present this capability currently doesn't have a way to
correctly set the data path. This add the support to set orientation via
a simple GPIO.
Signed-off-by: Xu Yang <xu.yang_2@nxp.com>
---
Changes in v2:
- return early in tcpci_set_orientation() if using gpio method
---
drivers/usb/typec/tcpm/tcpci.c | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)
diff --git a/drivers/usb/typec/tcpm/tcpci.c b/drivers/usb/typec/tcpm/tcpci.c
index 2a951c585e92..b930c8b1d2f7 100644
--- a/drivers/usb/typec/tcpm/tcpci.c
+++ b/drivers/usb/typec/tcpm/tcpci.c
@@ -7,6 +7,7 @@
#include <linux/bitfield.h>
#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/i2c.h>
@@ -42,6 +43,7 @@ struct tcpci {
struct tcpc_dev tcpc;
struct tcpci_data *data;
+ struct gpio_desc *orientation_gpio;
};
struct tcpci_chip {
@@ -316,6 +318,10 @@ static int tcpci_set_orientation(struct tcpc_dev *tcpc,
struct tcpci *tcpci = tcpc_to_tcpci(tcpc);
unsigned int reg;
+ if (tcpci->orientation_gpio)
+ return gpiod_set_value_cansleep(tcpci->orientation_gpio,
+ orientation == TYPEC_ORIENTATION_NORMAL ? 0 : 1);
+
switch (orientation) {
case TYPEC_ORIENTATION_NONE:
/* We can't put a single output into high impedance */
@@ -903,6 +909,7 @@ EXPORT_SYMBOL_GPL(tcpci_unregister_port);
static int tcpci_probe(struct i2c_client *client)
{
struct tcpci_chip *chip;
+ struct gpio_desc *orient_gpio = NULL;
int err;
u16 val = 0;
@@ -931,12 +938,23 @@ static int tcpci_probe(struct i2c_client *client)
if (err < 0)
return err;
+ if (err == 0) {
+ orient_gpio = devm_gpiod_get_optional(&client->dev, "orientation",
+ GPIOD_OUT_LOW);
+ if (IS_ERR(orient_gpio))
+ return dev_err_probe(&client->dev, PTR_ERR(orient_gpio),
+ "unable to acquire orientation gpio\n");
+ err = orient_gpio ? 1 : 0;
+ }
+
chip->data.set_orientation = err;
chip->tcpci = tcpci_register_port(&client->dev, &chip->data);
if (IS_ERR(chip->tcpci))
return PTR_ERR(chip->tcpci);
+ chip->tcpci->orientation_gpio = orient_gpio;
+
err = devm_request_threaded_irq(&client->dev, client->irq, NULL,
_tcpci_irq,
IRQF_SHARED | IRQF_ONESHOT,
--
2.34.1
From: Frank Li (AI-BOT) <frank.li@nxp.com> > + err = orient_gpio ? 1 : 0; AI: Redundant ternary. Use `err = !!orient_gpio;` or `err = orient_gpio != NULL;` for clarity. Frank
On Thu, Mar 12, 2026 at 03:12:41PM -0400, Frank Li wrote: > From: Frank Li (AI-BOT) <frank.li@nxp.com> > > > > + err = orient_gpio ? 1 : 0; > > AI: Redundant ternary. Use `err = !!orient_gpio;` or `err = orient_gpio != NULL;` > for clarity. It was "err = !!orient_gpio;" in V1. But the AI-BOT complained about it. I will return it back in V3. Thanks, Xu Yang
On Thu, Mar 12, 2026 at 03:12:41PM -0400, Frank Li wrote: > From: Frank Li (AI-BOT) <frank.li@nxp.com> > > > > + err = orient_gpio ? 1 : 0; > > AI: Redundant ternary. Use `err = !!orient_gpio;` or `err = orient_gpio != NULL;` > for clarity. Better yet, never use ? :, just spell it out with a real if statement. And ick, do NOT do that last option, go kick your AI bot, that's horrid. thanks, greg k-h
On Thu, Mar 12, 2026 at 09:27:09PM +0100, Greg Kroah-Hartman wrote: > On Thu, Mar 12, 2026 at 03:12:41PM -0400, Frank Li wrote: > > From: Frank Li (AI-BOT) <frank.li@nxp.com> > > > > > > > + err = orient_gpio ? 1 : 0; > > > > AI: Redundant ternary. Use `err = !!orient_gpio;` or `err = orient_gpio != NULL;` > > for clarity. > > Better yet, never use ? :, just spell it out with a real if statement. > And ick, do NOT do that last option, go kick your AI bot, that's horrid. last option is bad, I have not remove it when I review AI's result because I think first one look good. I generally just remove whole sentense from AI. Frank > > thanks, > > greg k-h
© 2016 - 2026 Red Hat, Inc.