Hi Matthew,
於 三，2010-04-28 於 22:43 +0100，Matthew Garrett 提到：
> On Thu, Apr 29, 2010 at 05:40:34AM +0800, joeyli wrote:
>
> > It causes userland application (e.g. OSD) need wait 5 seconds safely
> > time before show up OSD when user press wifi/bluetooth function key.
> > Why rfkill interface doesn't have show() hook can modify the rfkill
> > state immediately, don't need wait 5 seconds?
>
> Because that implies that userspace has to poll the device in order to
> get any updates. The intention is that the rfkill interface be event
> driven. Is there no way to get an event in the driver when the switch
> status is changed? See dell-laptop.c and hp-wmi.c for examples of this.
>
Yes, I saw the hp-wmi.c do that:
} else if (eventcode == 0x5) {
if (wifi_rfkill)
rfkill_set_sw_state(wifi_rfkill,
hp_wmi_wifi_state());
if (bluetooth_rfkill)
rfkill_set_sw_state(bluetooth_rfkill,
hp_wmi_bluetooth_state());
if (wwan_rfkill)
rfkill_set_sw_state(wwan_rfkill,
hp_wmi_wwan_state());
} else
And I want do the same thing in msi-laptop driver, but MSI BIOS send
keycode but not wmi event. So, I think I will reference the
rfkill-input.c to add a simply input handler to msi-laptop driver.
When it capture the RFKILL_TYPE_WLAN/RFKILL_TYPE_BLUETOOTH, it will
query the -REAL- hardware block then update rfkill state.
Do you think it's good?
Thank's
Joey Lee
--
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
