Hi,
On Sat, May 19, 2018, 14:21 Aditya Xavier <[email protected]> wrote: > Hi Michał / Łukasz, > > Were you able to identify the issue ? Do let me know if you need any > further testing from my end. > We found one issue. Not yet PR bit you can apply patch for you to test ble_adv_copy_to_ext_param(struct ble_gap_ext_adv_params *ext_param, const struct ble_gap_adv_params *param) @@ -510,6 +522,7 @@ ble_adv_copy_to_ext_param(struct ble_gap_ext_adv_params *ext_param, ext_param->itvl_min = param->itvl_min; ext_param->channel_map = param->channel_map; ext_param->high_duty_directed = param->high_duty_cycle; + ext_param->own_addr_type = g_mesh_addr_type; } This is not yet solving the issue but you should see adv going out from Device which uses BLE_EXT_ADV and uses non public address. > Also, do you recommend I submit a new Issue on Github for the same ? > Please do it. \Łukasz > From my testing its pretty apparent that BLE_EXT_ADV does not allow > bt_mesh_model_send ( mesh_init.c Line 155 ) to work anymore. > > Regarding the issue of Device B not receiving messages till around 15-50 > attempts, I believe it requires further deep dive. > > Am under the assumption that BLE_EXT_ADV does not actually require BLE to > be used and just Mesh to be configured. > > Do let me know if there are any issues in the code / my thought process. > > Thanks, > Aditya Xavier. > > > > > On 19-May-2018, at 12:23 PM, Aditya Xavier <[email protected]> wrote: > > > > Hi Michał, > > > > Sorry fo the confusion. I have done some more testing on the same, > please find the test results in the xl file. > > > > Please note, the test results are of Device A; wherein the Device B is > kept in the same state ( with all the mentioned flags turned off) > > > > Also, test cases 2 - 5 have an issue wherein right after loading the > firmware Device A would receive Messages from Dev B. > > > > However, Device A would not be able to send messages to Dev B, till > around 15-50 attempts. A restart of Dev B helps. > > > > The initial delay to receive messages was what I thought not working > earlier. > > <Book1.xlsx> > > > > Thanks, > > Aditya Xavier. > > > > > >> On 18-May-2018, at 6:27 PM, Michał Narajowski < > [email protected]> wrote: > >> > >> Hi Aditya, > >> > >> BLE_ROLE_BROADCASTER should not have an impact on this. There is only > one > >> place in the code where this is used: > >> > >> include/nimble/nimble_opt_auto.h:37: #define NIMBLE_BLE_ADVERTISE > >> \ > >> (MYNEWT_VAL(BLE_ROLE_BROADCASTER) || MYNEWT_VAL(BLE_ROLE_PERIPHERAL)) > >> > >> Best regards > >> Michał Narajowski > >> > >> pt., 18 maj 2018 o 14:02 Aditya Xavier <[email protected]> > napisał(a): > >> > >>> Hi Michał, > >>> > >>> A correction. > >>> > >>> It seems its out of the two BLE_ROLE* flags, its only > BLE_ROLE_BROADCASTER > >>> that has an adverse effect. > >>> Enabling BLE_ROLE_BROADCASTER stops the device from receiving messages. > >>> > >>> Thanks, > >>> Aditya Xavier. > >>> > >>> > >>> > >>> > >>>> On 18-May-2018, at 3:44 PM, Aditya Xavier <[email protected]> > wrote: > >>>> > >>>> Hi Michał, > >>>> > >>>> Yes, that is what you should be seeing, because the message is being > >>> sent to the GROUP; originator receives it as well. > >>>> If you change it to the destination / target address, you wouldn’t see > >>> Received. > >>>> > >>>> To test this, you would require two devices. > >>>> > >>>> Assuming Device A and B. > >>>> > >>>> With all three flags disabled and unique address ( node_address in > >>> mesh_init.c Line 28). > >>>> Device A ( button Pressed ) -> Device B should say in the Log > Received. > >>> ( Model callback method - mesh_init.c Line 78 ) > >>>> Device B ( button Pressed ) -> Device A should say in the Log > Received. > >>>> > >>>> With all three flags enabled and unique node_address. > >>>> Device A ( button Pressed )-> Device B will not receive any message. > >>>> Device B ( button Pressed )-> Device A will not receive any message. > >>>> > >>>> With only BLE_ROLE* flags disabled on Device B and all three flags > >>> disabled on Device B, and unique node_address. > >>>> Device A ( button Pressed )-> Device B should say in the Log Received. > >>>> Device B ( button Pressed )-> Device A will not receive any message. > >>>> > >>>> Thanks, > >>>> Aditya Xavier. > >>>> > >>>> > >>>>> On 18-May-2018, at 3:23 PM, Michał Narajowski < > >>> [email protected]> wrote: > >>>>> > >>>>> Hi Aditya, > >>>>> > >>>>> I enabled these flags: > >>>>> > >>>>> BLE_ROLE_BROADCASTER: 1 > >>>>> BLE_ROLE_PERIPHERAL: 1 > >>>>> BLE_EXT_ADV: 1 > >>>>> > >>>>> And this is what i see after pushing the button a few times: > >>>>> > >>>>> 045120 #mesh-onoff STATUS > >>>>> 045121 #mesh-onoff STATUS: Sent ! > >>>>> 045123 Received > >>>>> 045263 #mesh-onoff STATUS > >>>>> 045264 #mesh-onoff STATUS: Sent ! > >>>>> 045266 Received > >>>>> 045402 #mesh-onoff STATUS > >>>>> 045402 #mesh-onoff STATUS: Sent ! > >>>>> 045404 Received > >>>>> 045535 #mesh-onoff STATUS > >>>>> 045536 #mesh-onoff STATUS: Sent ! > >>>>> 045538 Received > >>>>> 046559 #mesh-onoff STATUS > >>>>> 046559 #mesh-onoff STATUS: Sent ! > >>>>> 046561 Received > >>>>> 046601 #mesh-onoff STATUS > >>>>> 046602 #mesh-onoff STATUS: Sent ! > >>>>> 046604 Received > >>>>> 046627 #mesh-onoff STATUS > >>>>> 046628 #mesh-onoff STATUS: Sent ! > >>>>> 046630 Received > >>>>> 046656 #mesh-onoff STATUS > >>>>> 046656 #mesh-onoff STATUS: Sent ! > >>>>> 046658 Received > >>>>> > >>>>> > >>>>> > >>>>> Is this what I should see? What are your symptoms? > >>>>> > >>>>> Best regards > >>>>> Michał Narajowski > >>>>> > >>>>> pt., 18 maj 2018 o 11:47 Aditya Xavier <[email protected]> > >>> napisał(a): > >>>>> > >>>>>> Hi Łukasz, > >>>>>> > >>>>>> Disabling only the following flags in syscfg.yml allows the device > to > >>>>>> receive but not send mesh messages. > >>>>>> > >>>>>> BLE_ROLE_BROADCASTER: 1 > >>>>>> BLE_ROLE_PERIPHERAL: 1 > >>>>>> > >>>>>> Disabling BLE_EXT_ADV: 1, flag allows the device to send and receive > >>> mesh > >>>>>> messages. > >>>>>> > >>>>>> And as I said earlier enabling all three of them, does not allow the > >>>>>> device to send / receive mesh messages. > >>>>>> > >>>>>> It is quite possible its a mistake on my end. Would be grateful if > you > >>> let > >>>>>> me know what am I doing wrong :) > >>>>>> > >>>>>> Thanks, > >>>>>> Aditya Xavier. > >>>>>> > >>>>>> > >>>>>>> On 18-May-2018, at 2:59 PM, Aditya Xavier <[email protected]> > >>> wrote: > >>>>>>> > >>>>>>> Hi Łukasz, > >>>>>>> > >>>>>>> Am actually sending it to the Group Address. > >>>>>>> > >>>>>>> In main.c :- Line 27 > >>>>>>> void button_cb(struct os_event *ev) > >>>>>>> { > >>>>>>> mesh_msg_send(MODEL_ID_CBOR_ACTION, GROUP_ADDR, "HELLO WORLD", 12); > >>>>>>> > >>>>>>> } > >>>>>>> > >>>>>>> And its relevant method :- > >>>>>>> > >>>>>>> In mesh_init.c :- Line 138. > >>>>>>> void mesh_msg_send(uint16_t model_id, uint16_t target_address, char > >>>>>> *tmsg, uint16_t tlen) > >>>>>>> { > >>>>>>> struct os_mbuf *msg = NET_BUF_SIMPLE(10 + tlen); > >>>>>>> struct bt_mesh_msg_ctx ctx = { > >>>>>>> .net_idx = net_idx, > >>>>>>> .app_idx = app_idx, > >>>>>>> // .addr = node_address, > >>>>>>> // .addr = GROUP_ADDR, > >>>>>>> .addr = target_address, > >>>>>>> .send_ttl = BT_MESH_TTL_MAX, > >>>>>>> }; > >>>>>>> > >>>>>>> > >>>>>>> > >>>>>>> Which I registered here :- > >>>>>>> In mesh_init.c :- Line 129. > >>>>>>> * Add model subscription */ > >>>>>>> bt_mesh_cfg_mod_sub_add_vnd(net_idx, node_address, node_address, > >>>>>> GROUP_ADDR, > >>>>>>> MOD_LF, CID_VENDOR, NULL); > >>>>>>> > >>>>>>> Also, I believe the mesh_msg_send is correctly because it works > >>>>>> correctly when I switch off the following flags. > >>>>>>> > >>>>>>> BLE_ROLE_BROADCASTER: 1 > >>>>>>> BLE_ROLE_PERIPHERAL: 1 > >>>>>>> BLE_EXT_ADV: 1 > >>>>>>> > >>>>>>> Please do correct me if am wrong. > >>>>>>> > >>>>>>> Thanks, > >>>>>>> Aditya Xavier. > >>>>>>> > >>>>>>>> On 18-May-2018, at 2:53 PM, Łukasz Rymanowski < > >>>>>> [email protected]> wrote: > >>>>>>>> > >>>>>>>> Hi Aditya, > >>>>>>>> > >>>>>>>> Sending to destination with same address as source will result in > >>>>>> sending > >>>>>>>> msg to source. It will not go into the air. > >>>>>>>> > >>>>>>>> Best > >>>>>>>> Łukasz > >>>>>>>> > >>>>>>>> On Thu, 17 May 2018 at 12:21, Aditya Xavier <[email protected]> > >>>>>> wrote: > >>>>>>>> > >>>>>>>>> Hi Łukasz, > >>>>>>>>> > >>>>>>>>> Sorry for the late reply. > >>>>>>>>> > >>>>>>>>> I created a test application for BLE + MESH co-existence, to test > >>> the > >>>>>>>>> functionality and your patch. > >>>>>>>>> > >>>>>>>>> The following are my observations. > >>>>>>>>> > >>>>>>>>> 1. On enabling any of the following flags, MESH is unable to > >>> send > >>>>>> / > >>>>>>>>> receive messages don’t work. (Using mesh_model_send) > >>>>>>>>> > >>>>>>>>> BLE_ROLE_BROADCASTER: 1 > >>>>>>>>> BLE_ROLE_PERIPHERAL: 1 > >>>>>>>>> BLE_EXT_ADV: 1 > >>>>>>>>> > >>>>>>>>> 2. Incase using the same Node_Address on two devices, send/ > >>>>>> receive > >>>>>>>>> messages don't work. ( This might be as per protocol > specifications, > >>>>>> am not > >>>>>>>>> aware) > >>>>>>>>> > >>>>>>>>> You can try the application to verify the same. > >>>>>>>>> 1. Do remember the node_address value should be unique. You > can > >>>>>>>>> change the same in mesh_init.c > >>>>>>>>> 2. You can enable BLE / mesh by changing the relevant value > in > >>>>>>>>> headers.h ( ble_enable and mesh_enable ) > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> Please let me know your findings and in case of any mistake on my > >>> part, > >>>>>>>>> please let me know. > >>>>>>>>> > >>>>>>>>> Thanks for looking into it. > >>>>>>>>> > >>>>>>>>> Thanks, > >>>>>>>>> Aditya Xavier. > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>>> On 15-May-2018, at 7:59 PM, Łukasz Rymanowski < > >>>>>>>>> [email protected]> wrote: > >>>>>>>>>> > >>>>>>>>>> Hi Aditya, > >>>>>>>>>> > >>>>>>>>>> On Sat, 12 May 2018 at 08:23, Aditya Xavier < > [email protected]> > >>>>>> wrote: > >>>>>>>>>> > >>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>> > >>>>>>>>>>> Thanks for looking into it. > >>>>>>>>>>> > >>>>>>>>>>> As per the documentation, > >>>>>>>>>>> > >>>>>>>>>>> To make use of it application needs to set BLE_EXT_ADV and > >>>>>>>>>>> BLE_MULTI_ADV_INSTANCES at least to 1. Mesh will use last > >>> available > >>>>>>>>>>> advertising instance for PB-ADV and second to last for PB-GATT > and > >>>>>>>>>>> PROXY > >>>>>>>>>>> > >>>>>>>>>>> What would happen when the Application is not utilizing PB-ADV > / > >>>>>>>>> PB-GATT ? > >>>>>>>>>>> i.e. provisioned Mesh Instance co-existing with a BLE instance > ? > >>>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> Ah this is error in the documentation. Instead of PB_ADV/GATT It > >>>>>> should > >>>>>>>>> be > >>>>>>>>>> ADV bearer and GATT bearer. Actually also name of #defines > >>>>>>>>>> BLE_ADV_PB_ADV/GATT_IDX is not good. Thanks for pointing it out. > >>> Would > >>>>>>>>> you > >>>>>>>>>> mind to send PR for this? > >>>>>>>>>> > >>>>>>>>>> In general, when BLE_EXT_ADV is enabled and application wants to > >>> use > >>>>>> BLE > >>>>>>>>>> instance and Mesh, then > >>>>>>>>>> a) BLE_MULTI_ADV_INSTANCES shall be set at least to 1 if ADV > >>> bearer > >>>>>> is > >>>>>>>>> used > >>>>>>>>>> b) BLE_MULTI_ADV_INSTANCES shall be set at least to 2 if ADV and > >>> GATT > >>>>>>>>>> bearer is used > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>>> Trying to understand whether this would require me to enable > >>> PB-ADV / > >>>>>>>>>>> PB-GATT because my application uses neither. > >>>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> There should be no need for this. > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>>> > >>>>>>>>>>> Similarly, please correct me if am wrong but I would need to > set > >>>>>>>>>>> BLE_MULTI_ADV_INSTANCES to 2 to have MESH and BLE coexist. > >>>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> No, 1 is enough. It is because BLE_MULTI_ADV_INSTANCES stays > for > >>> an > >>>>>>>>>> additional instances. There is always one available. > >>>>>>>>>> Actually I understand that name 'BLE_MULTI_ADV_INSTANCES' > might be > >>>>>>>>>> confusing. Hopefully decryption of this flag in > nimble/syscfg.yml > >>>>>> makes > >>>>>>>>> it > >>>>>>>>>> clear. > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>>> > >>>>>>>>>>> Thanks, > >>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>> > >>>>>>>>>>> > >>>>>>>>>> Best > >>>>>>>>>> Łukasz > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>>>> On 10-May-2018, at 1:16 PM, Łukasz Rymanowski < > >>>>>>>>>>> [email protected]> wrote: > >>>>>>>>>>>> > >>>>>>>>>>>> Hi, > >>>>>>>>>>>> > >>>>>>>>>>>> We found the issue which was related to latest controller > >>> changes. > >>>>>>>>>>>> Basically controller does not allow now to mix legacy HCI with > >>> ext > >>>>>> adv > >>>>>>>>>>> HCI > >>>>>>>>>>>> which of course is good. The PR > >>>>>>>>>>>> https://github.com/apache/mynewt-nimble/pull/8 is fixed now. > >>>>>>>>>>>> I removed RFC and I plan to merge it soon. > >>>>>>>>>>>> > >>>>>>>>>>>> @Aditya - feedback very welcome. > >>>>>>>>>>>> > >>>>>>>>>>>> \Łukasz > >>>>>>>>>>>> > >>>>>>>>>>>> On Mon, 7 May 2018 at 13:41, Łukasz Rymanowski < > >>>>>>>>>>>> [email protected]> wrote: > >>>>>>>>>>>> > >>>>>>>>>>>>> Hi Aditya, > >>>>>>>>>>>>> > >>>>>>>>>>>>> Sorry for late answer. > >>>>>>>>>>>>> > >>>>>>>>>>>>> Could you please provide btmon logs along with console logs > so > >>> we > >>>>>> can > >>>>>>>>>>> help > >>>>>>>>>>>>> you to analyze what is going on? > >>>>>>>>>>>>> Here is instruction how to get btmon logs: > >>>>>>>>>>>>> https://www.codecoup.pl/blog/support-for-btmon-in-mynewt/ > >>>>>>>>>>>>> > >>>>>>>>>>>>> Thanks and best regards > >>>>>>>>>>>>> Łukasz > >>>>>>>>>>>>> > >>>>>>>>>>>>> > >>>>>>>>>>>>> On Fri, 27 Apr 2018 at 11:48, Aditya Xavier < > >>> [email protected]> > >>>>>>>>>>> wrote: > >>>>>>>>>>>>> > >>>>>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>>>>> > >>>>>>>>>>>>>> Any update on it? Were you able to check this ? > >>>>>>>>>>>>>> > >>>>>>>>>>>>>> Bt_mesh_model_send does not work after enabling > BLE_EXT_ADV.. > >>>>>>>>>>>>>> > >>>>>>>>>>>>>> Or, can you give me a sample code where it works so that I > can > >>>>>> check > >>>>>>>>> if > >>>>>>>>>>>>>> there is something am doing wrong. > >>>>>>>>>>>>>> > >>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>>>>> > >>>>>>>>>>>>>> > >>>>>>>>>>>>>>> On 20-Apr-2018, at 3:56 PM, Aditya Xavier < > >>> [email protected]> > >>>>>>>>>>> wrote: > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> Thanks, was able to build btshell + blemesh into nrf52832. > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> I think I found an issue with regards to bt_mesh_model_send > >>>>>> method. > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> It seems bt_mesh_model_send is no longer working. > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> In my test app, I have remove all ble code, and using only > >>>>>> blemesh ( > >>>>>>>>>>>>>> auto provisioning ) and send a message to another device > over > >>> mesh > >>>>>>>>> on a > >>>>>>>>>>>>>> press of a button. > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> When I disable BLE_EXT_ADV and BLE_MULTI_ADV_INSTANCES, it > >>> works. > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> Can you try and confirm it works ? > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>>> Aditya Xavier > >>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> On 19-Apr-2018, at 2:00 PM, Łukasz Rymanowski < > >>>>>>>>>>>>>> [email protected] <mailto: > >>>>>> [email protected] > >>>>>>>>>>> > >>>>>>>>>>>>>> wrote: > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> Hi Aditya, > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> I suggest to hack around flash map instead of removing > code. > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> I would do something like this (hopefully calculations are > >>> OK). > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> +++ b/hw/bsp/nrf52dk/bsp.yml > >>>>>>>>>>>>>>>> @@ -41,11 +41,11 @@ bsp.flash_map: > >>>>>>>>>>>>>>>> FLASH_AREA_IMAGE_0: > >>>>>>>>>>>>>>>> device: 0 > >>>>>>>>>>>>>>>> offset: 0x00008000 > >>>>>>>>>>>>>>>> - size: 232kB > >>>>>>>>>>>>>>>> + size: 462kB > >>>>>>>>>>>>>>>> FLASH_AREA_IMAGE_1: > >>>>>>>>>>>>>>>> device: 0 > >>>>>>>>>>>>>>>> - offset: 0x00042000 > >>>>>>>>>>>>>>>> - size: 232kB > >>>>>>>>>>>>>>>> + offset: 0x0007B800 > >>>>>>>>>>>>>>>> + size: 2kB > >>>>>>>>>>>>>>>> FLASH_AREA_IMAGE_SCRATCH: > >>>>>>>>>>>>>>>> device: 0 > >>>>>>>>>>>>>>>> offset: 0x0007c000 > >>>>>>>>>>>>>>>> diff --git a/hw/bsp/nrf52dk/nrf52xxaa.ld > >>>>>>>>>>> b/hw/bsp/nrf52dk/nrf52xxaa.ld > >>>>>>>>>>>>>>>> index 9433e37fd..55e293da1 100644 > >>>>>>>>>>>>>>>> --- a/hw/bsp/nrf52dk/nrf52xxaa.ld > >>>>>>>>>>>>>>>> +++ b/hw/bsp/nrf52dk/nrf52xxaa.ld > >>>>>>>>>>>>>>>> @@ -17,7 +17,7 @@ > >>>>>>>>>>>>>>>> */ > >>>>>>>>>>>>>>>> MEMORY > >>>>>>>>>>>>>>>> { > >>>>>>>>>>>>>>>> - FLASH (rx) : ORIGIN = 0x00008000, LENGTH = 0x3a000 > >>>>>>>>>>>>>>>> + FLASH (rx) : ORIGIN = 0x00008000, LENGTH = 0x73800 > >>>>>>>>>>>>>>>> RAM (rwx) : ORIGIN = 0x20000000, LENGTH = 0x10000 > >>>>>>>>>>>>>>>> } > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> Make sure to rebuild boot and app and then try. > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> Best > >>>>>>>>>>>>>>>> Łukasz > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>> On 19 April 2018 at 07:30, Aditya Xavier < > >>> [email protected] > >>>>>>>>>>> <mailto: > >>>>>>>>>>>>>> [email protected]> <mailto:[email protected] <mailto: > >>>>>>>>>>>>>> [email protected]>>> wrote: > >>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> PFA the app I used to test Mesh using multi advertising > >>>>>> instances. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> I get the following error. > >>>>>>>>>>>>>>>>> [ts=275343728ssb, mod=4 level=3] adv_send: Advertising > >>> failed: > >>>>>>>>> err 3 > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> I had to comment out a lot of code to get it in a > flash-able > >>>>>> size. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 18-Apr-2018, at 1:13 PM, Aditya Xavier < > >>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Am using nrf52832, hence the problem of flash overflow. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Would create a test app, using btshell + blemesh + the > flags > >>>>>> which > >>>>>>>>>>> you > >>>>>>>>>>>>>>>>> recommended, and test again. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 18-Apr-2018, at 12:29 PM, Łukasz Rymanowski < > >>>>>>>>>>>>>>>>> [email protected] <mailto: > >>>>>>>>> [email protected] > >>>>>>>>>>>>> > >>>>>>>>>>>>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Aditya, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> If there is flash overflow consider removing some > features > >>> from > >>>>>>>>> the > >>>>>>>>>>>>>>>>> configuration. > >>>>>>>>>>>>>>>>> What HW are you using? We are running on nrf52840 > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> BTW There is no special application. It is btshell plus > >>> those 4 > >>>>>>>>>>> flags > >>>>>>>>>>>>>> ( > >>>>>>>>>>>>>>>>> BLE_EXT_ADV, BLE_MULTI_ADV_INSTANCES, BLE_MESH, > >>>>>> BLE_MESH_SHELL) . > >>>>>>>>>>> Of > >>>>>>>>>>>>>>>>> course you need my PR. I did not test it personally, > but I > >>>>>> know > >>>>>>>>> it > >>>>>>>>>>>>>> worked > >>>>>>>>>>>>>>>>> for Michal. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Please share your target configuration, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Best > >>>>>>>>>>>>>>>>> Łukasz > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 17 April 2018 at 12:04, Aditya Xavier < > >>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]> <mailto: > >>>>>>>>>>>>>>>>> [email protected] <mailto:[email protected]> > <mailto: > >>>>>>>>>>>>>> [email protected] <mailto:[email protected]>> < > >>>>>>>>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]> <mailto:[email protected] > >>> <mailto: > >>>>>>>>>>>>>> [email protected]>>>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Been trying to join both blemesh_shell and bt_shell, but > >>> there > >>>>>> is > >>>>>>>>> a > >>>>>>>>>>>>>>>>> problem of flash overflow. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Created another app, which basically is btshell and some > >>>>>> portions > >>>>>>>>> of > >>>>>>>>>>>>>>>>> blemesh, but that didn’t work. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Is it possible for you to share a sample / test app ? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 10-Apr-2018, at 1:09 PM, Łukasz Rymanowski < > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> [email protected] <mailto: > >>>>>>>>> [email protected] > >>>>>>>>>>>> > >>>>>>>>>>>>>> <mailto:[email protected] <mailto: > >>>>>>>>>>>>>> [email protected]>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Michał, Aditya, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> I just upload a new version of PR: > >>>>>>>>>>>>>>>>> https://github.com/apache/mynewt-nimble/pull/8 < > >>>>>>>>>>>>>> https://github.com/apache/mynewt-nimble/pull/8> < > >>>>>>>>>>>>>> https://github.com/apache/mynewt-nimble/pull/8 < > >>>>>>>>>>>>>> https://github.com/apache/mynewt-nimble/pull/8>> > >>>>>>>>>>>>>>>>> It contains fixes for the problem mentioned above, > however > >>>>>>>>> solution > >>>>>>>>>>> is > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> bit > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> different from what Michał suggested. > >>>>>>>>>>>>>>>>> @MIchał, could you take a look? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> @Aditya, Could you be able to test it and give us a > >>> feedback on > >>>>>>>>>>> this? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Best > >>>>>>>>>>>>>>>>> Łukasz > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 6 April 2018 at 14:08, Michał Narajowski < > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> [email protected] <mailto: > >>>>>>>>> [email protected] > >>>>>>>>>>>> > >>>>>>>>>>>>>> <mailto:[email protected] <mailto: > >>>>>>>>>>>>>> [email protected]>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Aditya, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Mesh is using Adv extensions under the hood if you have > >>>>>> Łukasz's > >>>>>>>>>>> patch > >>>>>>>>>>>>>>>>> and enable BLE_EXT_ADV and set BLE_MULTI_ADV_INSTANCES > to at > >>>>>> least > >>>>>>>>>>> 1. > >>>>>>>>>>>>>>>>> Blemesh_shell has a command "init" which initializes mesh > >>> stack > >>>>>>>>> and > >>>>>>>>>>>>>>>>> starts advertising Unprovisioned Mesh Beacon. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> I tested this now and I noticed a bug. Here is a patch > for > >>> that > >>>>>>>>> bug: > >>>>>>>>>>>>>>>>> https://pastebin.com/gbyX8H56 < > >>> https://pastebin.com/gbyX8H56> > >>>>>> < > >>>>>>>>>>>>>> https://pastebin.com/gbyX8H56 < > https://pastebin.com/gbyX8H56>> > >>>>>>>>>>>>>>>>> Please apply it on top of Łukasz's branch. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hope that helps. Let us know how it works for you. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> BR, > >>>>>>>>>>>>>>>>> Michał > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> 2018-04-06 11:09 GMT+02:00 Aditya Xavier < > >>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]> <mailto:[email protected] > >>> <mailto: > >>>>>>>>>>>>>> [email protected]>>>: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Michał / Łukasz, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> I have been trying to understand the blemesh_shell, and I > >>> fail > >>>>>> to > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> understand how / where it is using the Advertisement > >>>>>> extensions. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Basically, could you point me towards the difference if I > >>> need > >>>>>> to > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> implement, in order to use blemesh instead. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> From what I gathered / understood after going through the > >>> code > >>>>>> is > >>>>>>>>>>> that > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> blemesh_shell basically allows various functions to be > >>>>>> triggered > >>>>>>>>>>>>>> through > >>>>>>>>>>>>>>>>> shell commands. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 03-Apr-2018, at 3:54 PM, Michał Narajowski < > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> [email protected] <mailto: > >>>>>>>>> [email protected] > >>>>>>>>>>>> > >>>>>>>>>>>>>> <mailto:[email protected] <mailto: > >>>>>>>>>>>>>> [email protected]>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Aditya, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Please set BLE_MESH: 1 and BLE_MESH_SHELL: 1 and you > should > >>> be > >>>>>>>>> able > >>>>>>>>>>> to > >>>>>>>>>>>>>>>>> use both btshell and mesh shell. Let us know how that > works > >>> for > >>>>>>>>> you. > >>>>>>>>>>>>>>>>> [email protected] <mailto:[email protected]> <mailto: > >>>>>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]>> <mailto:[email protected] <mailto: > >>>>>>>>>>> [email protected]> > >>>>>>>>>>>>>> <mailto:[email protected] <mailto:[email protected]>> < > >>>>>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]> <mailto:[email protected] <mailto: > >>>>>>>>>>> [email protected] > >>>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Best regards > >>>>>>>>>>>>>>>>> Michał > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> 2018-04-03 7:56 GMT+02:00 Aditya Xavier < > >>> [email protected] > >>>>>>>>>>> <mailto: > >>>>>>>>>>>>>> [email protected]> <mailto:[email protected] <mailto: > >>>>>>>>>>>>>> [email protected]>> <mailto: > >>>>>>>>>>>>>>>>> [email protected] <mailto:[email protected]> > <mailto: > >>>>>>>>>>>>>> [email protected] <mailto:[email protected]>> < > >>>>>>>>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]> <mailto:[email protected] > >>> <mailto: > >>>>>>>>>>>>>> [email protected]>>>>>: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Any pointers, as to what needs to be implemented from the > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> blemesh_shell app ? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 02-Apr-2018, at 11:34 PM, Łukasz Rymanowski < > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> [email protected] <mailto: > >>>>>>>>> [email protected] > >>>>>>>>>>>> > >>>>>>>>>>>>>> <mailto:[email protected] <mailto: > >>>>>>>>>>>>>> [email protected]>> <mailto: > >>>>>>>>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]> <mailto: > >>>>>>>>>>>>>> [email protected] <mailto: > >>>>>> [email protected] > >>>>>>>>>>> > >>>>>>>>>>>>>>>>> <[email protected] <mailto: > >>>>>>>>>>> [email protected]> > >>>>>>>>>>>>>> <mailto:[email protected] <mailto: > >>>>>>>>>>>>>> [email protected]>>>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Second thought > >>>>>>>>>>>>>>>>> [email protected] <mailto:[email protected]> <mailto: > >>>>>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]>> <mailto:[email protected] <mailto: > >>>>>>>>>>> [email protected]> > >>>>>>>>>>>>>> <mailto:[email protected] <mailto:[email protected]>> < > >>>>>> [email protected] > >>>>>>>>>>>>>> <mailto:[email protected]> <mailto:[email protected] <mailto: > >>>>>>>>>>> [email protected] > >>>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Aditya, > >>>>>>>>>>>>>>>>> Since I did not test it a lot, would it be possible to > give > >>> us > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> feedback how > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> it works for you? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Best > >>>>>>>>>>>>>>>>> Lukasz > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On Mon, Apr 2, 2018, 20:01 Łukasz Rymanowski < > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> [email protected]> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Aditya. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On Mon, Apr 2, 2018, 19:14 Aditya Xavier < > >>> [email protected]> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi Łukasz, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Is there anything special required to get this working > along > >>>>>> with > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> BLE ? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> For e.g. would the btshell app code for ADV_EXT work > along > >>> with > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> mesh with > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> the provided patches ? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> In addition to configuration mentioned in PR commit > message, > >>>>>> the > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> btshell > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> app would have to enable ble mesh and ble mesh shell > (check > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> blemesh_shell > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> app for that) > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Mesh and BLE seems to compile however, am currently > unable > >>> to > >>>>>> get > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Mesh > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> working. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Also, any reason why this was not accepted yet ? > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> People are busy with other stuff I guess. I think it > will be > >>>>>>>>> merged > >>>>>>>>>>>>>>>>> eventually. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>>>>>> Aditya Xavier. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Best > >>>>>>>>>>>>>>>>> Lukasz > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 20-Mar-2018, at 12:57 AM, Sterling Hughes < > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> [email protected]> wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> +1 - this is great, thanks Lukasz! > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> On 19 Mar 2018, at 6:36, Łukasz Rymanowski wrote: > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Hi All, > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> I saw people asking around about possibility to advertise > >>> with > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> non-mesh > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> data while mesh is running on Mynewt. Well this is > >>> possible to > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> do but > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> of > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> course it brings a lot of risk for mesh operations and > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> especially > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> for > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> friendship scenario. However I went ahead and added > support > >>> for > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> this in > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Mynewt and I'm interested in how it works for you. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> So here is a PR which makes use of multi instances from > BT5 > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Advertising > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> extensions and basically allows you to create additional > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> instances > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> which > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> contains non-mesh data. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> https://github.com/apache/mynewt-nimble/pull/8 > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Instructions on how to enable it you can find in the > commit > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> message. > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Comments are welcome > >>>>>>>>>>>>>>>>> > >>>>>>>>>>>>>>>>> Best > >>>>>>>>>>>>>>>>> Łukasz > >>>>>>>>>>>>>> > >>>>>>>>>>>>>> > >>>>>>>>>>> > >>>>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>> > >>>>>> > >>>>>> > >>>> > >>> > >>> > > > >
