Is there any difference between turning bluetooth on and off in the Settings window versus the top-right bar? Or how they receive/read the bluetooth status? Because the differences I see are that mine applies to both turning on and off and that it also affects the Settings window.

Since the top-right bar is behaving strangely with rfkill, switch or pressing "turn off", I didn't test it, but only the Settings window. I also didn't test it using my laptop airplane mode.

rfkill output and behaviours:
The ID for bluetooth device hci0 changes after execution of "rfkill block 1" or switch off. Also, hci0 reappears with a new ID (+1 the previous).


>>>rfkill (bluetooth is on)
ID TYPE      DEVICE              SOFT      HARD
 0 wlan      acer-wireless  unblocked unblocked
 1 bluetooth acer-bluetooth unblocked unblocked
 2 wlan      phy0           unblocked unblocked
21 bluetooth hci0           unblocked unblocked

>>>rfkill (bluetooth is off due to switch or rfkill block 1)
ID TYPE      DEVICE              SOFT      HARD
 0 wlan      acer-wireless  unblocked unblocked
 1 bluetooth acer-bluetooth   blocked unblocked
 2 wlan      phy0           unblocked unblocked

>>>rfkill (bluetooth is off due to rfkill block 53)
ID TYPE      DEVICE              SOFT      HARD
 0 wlan      acer-wireless  unblocked unblocked
 1 bluetooth acer-bluetooth unblocked unblocked
 2 wlan      phy0           unblocked unblocked
53 bluetooth hci0             blocked unblocked

If bluetooth is currently on and if I use the switch to turn it off and on again, or if bluetooth is currently off and I use the switch to turn it on, the delay mentioned in the bug description will appear as described.


ACER-BLUETOOTH DEVICE

If bluetooth is currently on: "rfkill block 1" will make Settings window instantly show as off while the switch stays on, it doesn't goes off until I change panels and come back.
If bluetooth is currently off due to "rfkill block 1": "rfkill unblock 1" will make Settings window show as on after 3-8s
If bluetooth is currently off due to "rfkill block 46": "rfkill unblock 1" won't work
If bluetooth is currently off due to switch: "rfkill unblock 1" will make Settings window and switch show as on after 6-8s (the command itself returns earlier, after 3s, and it is already updated with acer-bluetooth as unblocked)


HCI0 DEVICE

It should be noted that "rfkill block 46" followed by "rfkill unblock 46" didn't reproduce the delay, but kept the disparity between switch and state (Settings window)/"Off and turn off" (top-right bar).


If bluetooth is currently on: "rfkill block 46" will display the same behaviour as "rfkill block 1"
If bluetooth is currently off due to "rfkill block 1": I cannot use "rfkill unblock 46" as it doesn't exist and would return with ID 47.
If bluetooth is currently off due to "rfkill block 46": "rfkill unblock 46" will make Settings window show as on immediately.
If bluetooth is currently off due to switch: hci0 disappears from rfkill output.
-Unblocking via switch takes a lot of time, rfkill itself updates immediately to show acer-bluetooth as unblocked, but hci0 reappears only after the change takes effect in Settings window.
-Unblocking via "rfkill unblock 1" will update the Settings window after 8s (the command itself returns immediately, but the same behaviour is present (immediate for acer-bluetooth, delayed for hci0)).