From: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
These are spawned as children of the device that normally represents
the chip/hw block providing Type-C functionality. They don't have PM
callbacks, so it makes no sense for them to be taken into account
there.
Signed-off-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
---
I saw no regressions, but such changes are always a little scary..
Sending as RFT
---
drivers/usb/typec/mux.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/usb/typec/mux.c b/drivers/usb/typec/mux.c
index db5e4a4c0a99..080252f202ec 100644
--- a/drivers/usb/typec/mux.c
+++ b/drivers/usb/typec/mux.c
@@ -199,6 +199,8 @@ typec_switch_register(struct device *parent,
return ERR_PTR(ret);
}
+ device_set_pm_not_required(&sw_dev->dev);
+
ret = device_add(&sw_dev->dev);
if (ret) {
dev_err(parent, "failed to register switch (%d)\n", ret);
@@ -454,6 +456,8 @@ typec_mux_register(struct device *parent, const struct typec_mux_desc *desc)
return ERR_PTR(ret);
}
+ device_set_pm_not_required(&mux_dev->dev);
+
ret = device_add(&mux_dev->dev);
if (ret) {
dev_err(parent, "failed to register mux (%d)\n", ret);
---
base-commit: db7efce4ae23ad5e42f5f55428f529ff62b86fab
change-id: 20260409-topic-usb_mux_nopm-0e52479160c6
Best regards,
--
Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>