Hi Jakub, >>> --- >>> doc/mgmt-api.txt | 69 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ >>> 1 file changed, 69 insertions(+) >>> >>> diff --git a/doc/mgmt-api.txt b/doc/mgmt-api.txt >>> index 11e2490..1c78f1c 100644 >>> --- a/doc/mgmt-api.txt >>> +++ b/doc/mgmt-api.txt >>> @@ -2135,6 +2135,34 @@ Set Public Address Command >>> Invalid Index >>> >>> >>> +Start LE Filtered Device Scan >>> +======================= >>> + >>> + Command Code: 0x003A >>> + Controller Index: <controller id> >>> + Command Parameters: UUID (16 Octets) >>> + max_pathloss (1 octet) >>> + is_active (1 octet) >>> + >>> + Return Parameters: >>> + >>> + This command starts LE scan, and filter received advertisements: if >>> AD contains both TX power level and Service Solicitation, and UUID is >>> contained in Service Solicitation and computed pathloss is smaller >>> than max_pathloss parameter, then a Advertisement Received event will >>> be sent. >>> + >>> + Possible values for the is_active parameter: >>> + 0 Start passive scan >>> + 1 Start active scan >> >> is really like a one-shot scan? Or is this something you would be running all the time in the background? >> >> Fundamentally if it is one-shot thing then you always want to do active scan, but if it runs in the background then you only want to do passive scanning. > > I'm working on two projects, one require scanning up to 2 minutes, > second requires long term scanning in background. > I'd preffer to have scan type as argument, but having only passive > scan is also good option for me. I prefer to have these as separate commands. So one is actually a specialized active scanning to find a specific set of devices. Similar to Start Discovery. The other is to add an additional filter for our background scanning. Remember that this has to work in sync with the auto-connection logic that the kernel is running. If you are messing that up, then you might find the device you are looking for, but the HID mouse will not re-connect. >>> + >>> + This command can only be used when the controller is powered. >>> + >>> + This command generates a Command Complete event on success or failure. >>> + >>> + Possible errors: Busy >>> + Not Supported >>> + Invalid Parameters >>> + Not Powered >>> + Invalid Index >>> + >>> + >>> Command Complete Event >>> ====================== >>> >>> @@ -2851,3 +2879,44 @@ New Configuration Options Event >>> >>> This event indicates that one or more of the options for the >>> controller configuration has changed. >>> + >>> + >>> +Advertisement Received Event >>> +============================== >>> + >>> + Event Code: 0x0020 >>> + Controller Index: <controller id> >>> + Event Parameters: Address (6 Octets) >>> + Address_Type (1 Octet) >>> + RSSI (1 Octet) >>> + Flags (4 Octets) >>> + AD_Data_Length (2 Octets) >>> + AD_Data (0-65535 Octets) >> >> Device Found event not good enough here? Why would you need a new event with exactly the same parameters? > > Because Device Found event always contains EIR data, and mine would > sometimes contain only AD data. > Also this will be send for each advertisement received, not for first one only. Do not get confused because Device Found says EIR_Data. It will return EIR_Data for BR/EDR devices and AD_Data for LE devices. We just never bothered to find a more common name. We could easily send multiple Device Found events. Besides Start Discovery, we also have Add Device with Action 0x00 sending out Device Found events. So that would not be a problem. We just need to define the context properly. Regards Marcel -- To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
