Messages in this thread | | | From | Victor Ding <> | Date | Tue, 24 Jan 2023 18:46:13 +1100 | Subject | Re: [PATCH v2] platform/chrome: cros_ec_typec: allow deferred probe of switch handles |
| |
On Tue, Jan 24, 2023 at 5:57 PM Guenter Roeck <groeck@google.com> wrote: > > On Mon, Jan 23, 2023 at 10:16 PM Victor Ding <victording@chromium.org> wrote: > > > > `fwnode_typec_{retimer,mux,switch}_get()` could return `-EPROBE_DEFER`, > > which is called from `cros_typec_get_switch_handles`. When this happens, > > it does not indicate absence of switches; instead, it only hints that > > probing of switches should occur at a later time. > > > > Progagate `-EPROBE_DEFER` to upper layer logic so that they can re-try > > probing switches as a better time. > > > > Signed-off-by: Victor Ding <victording@chromium.org> > > --- > > > > Changes in v2: > > - Coverted switch-block to nested if-blocks. > > > > drivers/platform/chrome/cros_ec_typec.c | 26 ++++++++++++++++--------- > > 1 file changed, 17 insertions(+), 9 deletions(-) > > > > diff --git a/drivers/platform/chrome/cros_ec_typec.c b/drivers/platform/chrome/cros_ec_typec.c > > index 59de4ce01fab..580f3bdf091c 100644 > > --- a/drivers/platform/chrome/cros_ec_typec.c > > +++ b/drivers/platform/chrome/cros_ec_typec.c > > @@ -145,31 +145,37 @@ static int cros_typec_get_switch_handles(struct cros_typec_port *port, > > struct fwnode_handle *fwnode, > > struct device *dev) > > { > > + int ret = 0; > > + > > port->mux = fwnode_typec_mux_get(fwnode, NULL); > > if (IS_ERR(port->mux)) { > > - dev_dbg(dev, "Mux handle not found.\n"); > > + ret = PTR_ERR(port->mux); > > + dev_dbg(dev, "Mux handle not found: %d.\n", ret); > > goto mux_err; > > } > > > > port->retimer = fwnode_typec_retimer_get(fwnode); > > if (IS_ERR(port->retimer)) { > > - dev_dbg(dev, "Retimer handle not found.\n"); > > + ret = PTR_ERR(port->retimer); > > + dev_dbg(dev, "Retimer handle not found: %d.\n", ret); > > goto retimer_sw_err; > > } > > > > port->ori_sw = fwnode_typec_switch_get(fwnode); > > if (IS_ERR(port->ori_sw)) { > > - dev_dbg(dev, "Orientation switch handle not found.\n"); > > + ret = PTR_ERR(port->ori_sw); > > + dev_dbg(dev, "Orientation switch handle not found: %d\n", ret); > > goto ori_sw_err; > > } > > > > port->role_sw = fwnode_usb_role_switch_get(fwnode); > > if (IS_ERR(port->role_sw)) { > > - dev_dbg(dev, "USB role switch handle not found.\n"); > > + ret = PTR_ERR(port->role_sw); > > + dev_dbg(dev, "USB role switch handle not found: %d\n", ret); > > goto role_sw_err; > > } > > > > - return 0; > > + return ret; > > > > This change is pointless. Sure, I'll revert this line to `return 0` in the next revision. > > Guenter > > > role_sw_err: > > typec_switch_put(port->ori_sw); > > @@ -181,7 +187,7 @@ static int cros_typec_get_switch_handles(struct cros_typec_port *port, > > typec_mux_put(port->mux); > > port->mux = NULL; > > mux_err: > > - return -ENODEV; > > + return ret; > > } > > > > static int cros_typec_add_partner(struct cros_typec_data *typec, int port_num, > > @@ -423,9 +429,11 @@ static int cros_typec_init_ports(struct cros_typec_data *typec) > > } > > > > ret = cros_typec_get_switch_handles(cros_port, fwnode, dev); > > - if (ret) > > - dev_dbg(dev, "No switch control for port %d\n", > > - port_num); > > + if (ret) { > > + dev_dbg(dev, "No switch control for port %d, err: %d\n", port_num, ret); > > + if (ret == -EPROBE_DEFER) > > + goto unregister_ports; > > + } > > > > ret = cros_typec_register_port_altmodes(typec, port_num); > > if (ret) { > > -- > > 2.39.1.405.gd4c25cc71f-goog > >
| |