| T42 | <elros34> @edp_17 what doesn't work? | 00:00 |
|---|---|---|
| T42 | <edp_17> Well, not much works. After a reboot I can turn the bt on for a second (I can see the device name and visibility) but it turns off immediately. After that I cannot turn it back on. I have prepared some info for you: https://paste.ubuntu.com/p/GMxk29dG5c/ | 00:04 |
| T42 | <elros34> maybe first mask bluetooth-rfkill-event, reboot and make sure it doesn't start. Then figure out what do you need to use bluetooth. | 00:08 |
| T42 | <elros34> For sure you need "rfkill unblock all" and init it with brcm_patchram_plus at least once | 00:10 |
| Umeaboy | Hi! Is there something wrong with updating the rootfs version in the HADK? | 00:10 |
| Umeaboy | It's from 2018. | 00:10 |
| Umeaboy | TARBALL=ubuntu-trusty-20180613-android-rootfs.tar.bz2 | 00:10 |
| Umeaboy | I'm just curious to what the reason might be for using an older version of Ubuntu. | 00:11 |
| T42 | <elros34> I guess for compability with older android version. You don't have to use it | 00:11 |
| T42 | <edp_17> @elros34 : Okay, so I need this command 'ln -s /dev/null /etc/systemd/system/bluetooth-rfkill-event.service' then reboot? | 00:12 |
| T42 | <elros34> @edp_17 check whether init is enough to get bluetooth working at least once | 00:12 |
| Umeaboy | Trusty is a LTS version as it seems. | 00:12 |
| T42 | <elros34> @edp_17 should be enough, just make sure it doesn't start (pgrep -af bluetooth) | 00:13 |
| Umeaboy | I wonder why it's called Trusty from 2018 when Trusty was released back in 2014. | 00:15 |
| Umeaboy | https://wiki.ubuntu.com/Releases | 00:15 |
| Umeaboy | The Standard support for it has ended. | 00:15 |
| T42 | <edp_17> @elros34 : I have masked that service, rebooted and the 'pgrep -af bluetooth' returns with this: https://paste.ubuntu.com/p/kVp8vwTXSF/ | 00:19 |
| T42 | <elros34> If it's disabled then you don't have to pastebin it. Now run rfkill and init chip with brcm_patchram_plus | 00:30 |
| T42 | <edp_17> Do you meant start the bluetooth-rfkill-event.service? Then run manually the '/usr/sbin/brcm_patchram_plus --enable_hci --enable_lpm --no2bytes --tosleep 50000 --baudrate 3000000 --use_baudrate_for_download --patchram /system/bin/bcm4330B1.hcd --scopcm=0,0,0,0,0,0,0,3,3,0 /dev/ttySAC0' command? | 00:32 |
| T42 | <elros34> no, forget about any services. Just init it manually | 00:33 |
| T42 | <edp_17> I ran that command ('/usr/sbin/brcm_patchram_plus --enable_hci --enable_lpm --no2bytes --tosleep 50000 --baudrate 3000000 --use_baudrate_for_download --patchram /system/bin/bcm4330B1.hcd --scopcm=0,0,0,0,0,0,0,3,3,0 /dev/ttySAC0') and got 'Done setting line discipline' | 00:34 |
| T42 | <edp_17> And BT is on. | 00:35 |
| T42 | <elros34> Does it work correctly? | 00:35 |
| T42 | <edp_17> I successfully paired with my XA2. How can I test whether it is functional? | 00:36 |
| T42 | <elros34> so it works at least partially:) Now just disable bluetooth in menu and enable. Check whether it still works | 00:37 |
| T42 | <edp_17> I have connected to my bluetooth headset but the music still comes from the phone. :) | 00:38 |
| T42 | <elros34> lets focus on init and on/of first | 00:39 |
| T42 | <edp_17> I turned it off. Then waited a bit and turned it on again but it doesn't turn on. It is just blinking. | 00:40 |
| T42 | <elros34> check output of rfkill and hciconfig | 00:41 |
| T42 | <edp_17> https://paste.ubuntu.com/p/YmrtsQvGg2/ | 00:42 |
| T42 | <elros34> try some command like: echo " " > /dev/ttySAC0 or hciconfig hci0 up , maybe it's not neccessary to call brcm_patchram_plus every time you want to enable bluetooth | 00:46 |
| T42 | <edp_17> The 'echo " " > /dev/ttySAC0' doesn't return with the prompt. The 'hciconfig hci0 up' gives me: 'Can't get device info: No such device' and none of them turns bt on. | 00:48 |
| T42 | <edp_17> It looks like need to call brcm_patchram_plus to turn it on. | 00:49 |
| T42 | <elros34> maybe so execute it | 00:50 |
| T42 | <edp_17> I did and it turned bt on. | 00:50 |
| T42 | <elros34> So why it turns off immediately when using bluetooth-rfkill-event> If manual init works everytime then run udevadm monitor -p and check whether there is something unusual when you eanble and init bluetooth | 00:53 |
| T42 | <edp_17> I tried to pair with my bt keyboard. The phone sees it but cannot pair: 'Problem with pairing Coulr not pair the devices.' | 00:53 |
| T42 | <edp_17> I don't know. I compared the command I just ran against what the system runs with the event. The only difference that for the system one, there is a mac address in the call f0:a3:2f:8c:9f:7c (https://paste.ubuntu.com/p/GMxk29dG5c/) | 00:56 |
| T42 | <edp_17> While that is not there in the manual command: '/usr/sbin/brcm_patchram_plus --enable_hci --enable_lpm --no2bytes --tosleep 50000 --baudrate 3000000 --use_baudrate_for_download --patchram /system/bin/bcm4330B1.hcd --scopcm=0,0,0,0,0,0,0,3,3,0 /dev/ttySAC0' | 00:56 |
| T42 | <edp_17> But I am not sure that is an issue or not. | 00:56 |
| T42 | <elros34> you should add all parameters including bd_addr. What bout udevadm output? | 00:57 |
| T42 | <edp_17> I needed to stop the brcm_patchram_plus to be able to run the udevadm. I started 'udevadm monitor -p' then turned the bt on (tried): https://paste.ubuntu.com/p/8DkNYmZTs7/ | 01:01 |
| T42 | <edp_17> What I spotted in the udevadm is 'systemd-rfkill@rfkill0.service'. This service is broken: https://paste.ubuntu.com/p/M29ZNVMvqd/ | 01:03 |
| T42 | <elros34> RFKILL_STATE=0 means you disabled it. You were supposed to record what happens when you enable and init bluetooth | 01:03 |
| T42 | <edp_17> I didn't know that, I just ended the command and tap on the bluetooth text. π This should show the correct status: https://paste.ubuntu.com/p/rCTSvXfHpF/ | 01:05 |
| T42 | <elros34> the point is to figure out why bluetooth-rfkill-event fails if manuall on/off works, right | 01:07 |
| T42 | <edp_17> Yes. | 01:08 |
| T42 | <edp_17> But I couldn't work out. | 01:08 |
| T42 | <edp_17> Does that mac address make any difference? | 01:09 |
| T42 | <elros34> I don't think so | 01:09 |
| T42 | <edp_17> The two differences between the parameters when I run manually or when it runs by the system: | 01:11 |
| T42 | <edp_17> --enable_fork and --bd_addr f0:a3:2f:8c:9f:7c | 01:11 |
| T42 | <edp_17> Both of them are there when the system runs it, but I didn't include them for the manual run. | 01:12 |
| T42 | <elros34> maybe you could run blluetooth-rfkill-event manually with verbose mode | 01:15 |
| T42 | <edp_17> I have added those two parameters to my manual command and now that doesn't turn the bt on. Even if I remove them. π | 01:16 |
| T42 | <edp_17> Reboot... | 01:16 |
| T42 | <elros34> wait check whether patchram is still running in background | 01:16 |
| T42 | <edp_17> I did with ps ax, it wasn't running. | 01:18 |
| T42 | <elros34> so which option cause bt on fail? | 01:26 |
| T42 | <edp_17> When I included both the --enable_fork and --bd_addr f0:a3:2f:8c:9f:7c or only --bd_addr f0:a3:2f:8c:9f:7c to my manual command, the pairing failed. | 01:27 |
| T42 | <edp_17> When I removed both, the pairing worked. | 01:29 |
| T42 | <elros34> so --bd_addr f0:a3:2f:8c:9f:7c is culprit or both? | 01:29 |
| T42 | <edp_17> Well, I am not sure anymore because now I cannot pair without those two. | 01:30 |
| T42 | <edp_17> Why the bt is so difficult? π | 01:31 |
| T42 | <elros34> it works with small issue for most devices:P | 01:33 |
| T42 | <edp_17> Yeah, I know this one is an old crap. π | 01:34 |
| T42 | <elros34> so do you have correct init command which always work? Then make sure bluetooth-rfkill-event use same and grab verbose logs | 01:34 |
| T42 | <edp_17> I use the same what I used when we started few hours ago. And now, after a reboot, the command works again. | 01:35 |
| T42 | <edp_17> I guess the two additional parameter just messed up something but the reboot cleared that out. | 01:36 |
| T42 | <elros34> so use the same params in rfkil-event now and check how it works. btw from where did you get all these params like scopcm, tosleep, enable_lpm. Are you sure they are correct? | 01:37 |
| Umeaboy | elros34: The 5.1 part, how do I use that now ? I have already fully synced Lineage for this device. | 01:37 |
| T42 | <edp_17> Yeah, on that one you mentioned. In the '/etc/firmware/bcm4330b1.conf' there is no --enable_fork and --bd_addr f0:a3:2f:8c:9f:7c parameters, so I am not sure where it is getting them. | 01:37 |
| Umeaboy | The guide should have a "If you already have synced Lineage go here:" step. | 01:38 |
| T42 | <elros34> @edp_17 check sources and then change it in config: https://github.com/mer-hybris/bluetooth-rfkill-event/blob/master/bluetooth-rfkill-event/bluetooth_rfkill_event.c#L533 | 01:39 |
| T42 | <edp_17> @elros34 : (scopcm, tosleep, enable_lpm) I got them from the init script that you showed me a while ago. | 01:39 |
| T42 | <edp_17> Thanks I will change the source. | 01:40 |
| T42 | <elros34> no do not change source, just read it: boolean = g_key_file_get_boolean(config, "General", "fork", &err); | 01:40 |
| Umeaboy | 5.1 Checking out Source of the Android base tells you to setup for a hybris branch that hasn't been officially merged yet. | 01:40 |
| Umeaboy | Especially for hybris-17.1 | 01:40 |
| Umeaboy | And NO, the phone cannot be downgraded as far as I know even though the bootloader is unlocked.. | 01:41 |
| T42 | <elros34> Umeaboy if there are no hybris-17.1 patches then how do you want to build sailfish for that base? | 01:42 |
| T42 | <edp_17> @elros34 : By the way I have changed this line: https://github.com/mer-hybris/bluetooth-rfkill-event/blob/master/bluetooth-rfkill-event/bluetooth_rfkill_event.c#L78 | 01:42 |
| T42 | <edp_17> changed bcm43xx to bcm4330 | 01:43 |
| Umeaboy | elros34: There is one as I told you. It hasn't been merged yet: https://github.com/mer-hybris/hybris-patches/pull/3 | 01:44 |
| Umeaboy | I'm still waiting for it to be merged. | 01:44 |
| T42 | <elros34> why do you wait? If it works then just use it | 01:45 |
| Umeaboy | hybris-16.0 is not based of Android Q AFAIK. | 01:45 |
| Umeaboy | elros34: I'm no expert in how to transform that patch into a hybris-17.1 branch. | 01:46 |
| Umeaboy | Can you help me do that? I'm not telling you to do the job. Just guide me on how I can made the hybris-17.0 branch and use that. | 01:48 |
| T42 | <elros34> If you have all hybris-17 patches then execute hybris-patches/apply-patches.sh --mb like explained in faq: https://github.com/mer-hybris/hadk-faq#hybris-16-0 | 01:48 |
| T42 | <Rodrigosolari> I am thinking of creating a service, which starts a sh for the taking of logs, the detail is that how long should I put a delay for it to take the correct logs | 01:50 |
| T42 | <edp_17> @elros34 : Thanks for your help and good night. | 01:53 |
| T42 | <elros34> ee we have done nothing..:P | 01:53 |
| T42 | <Juan Pablo %lastname%> Fwd from page: Most times we get set backs and limitations by our own thoughts and decisions. Just like me but I have over come all those thoughts. And statistics has it that 90% of us people investing our investment are wrapped in fear and negativity. Until we all see the we begin to see the investment as an opportu | 05:22 |
| T42 | nity and a journey, we will end up failing and going back to the scratch. I want serious minded people to inbox Mrs Victoria now.. | 05:22 |
| T42 | <Juan Pablo %lastname%> @vakdez_1 | 05:22 |
| T42 | <Juan Pablo %lastname%> WHATSAPP +1 408-906-8137 | 05:22 |
| T42 | <Juan Pablo %lastname%> https://t.me/joinchat/AAAAAEcDyKryy9tKJNGRfQ | 05:22 |
| T42 | <Juan Pablo %lastname%> Fwd from page: Investing in binary options and forex trading with Mrs Victoria has been the best has ever happened to me, gaining financial freedom is a wonderful thing, and I started off small but now I can get whatever I want with my easy investment steps and payment returns. | 05:27 |
| T42 | <Juan Pablo %lastname%> WHATAPP π+1 408-906-8137 | 05:27 |
| T42 | <Juan Pablo %lastname%> @vakdez_1 | 05:27 |
| T42 | <Juan Pablo %lastname%> https://t.me/joinchat/AAAAAEcDyKryy9tKJNGRfQ | 05:27 |
| T42 | <birdzhang> git.merproject.org SSL certificate problem: certificate has expired | 05:42 |
| T42 | <birdzhang> and git.sailfishos.org will expire tomorrow | 05:48 |
| T42 | <Pepelooklikefrog> https://paste.ubuntu.com/p/gfTypZx3fk/ | 05:54 |
| T42 | <Pepelooklikefrog> Can someone help me? | 05:54 |
| T42 | <neochapay> @Pepelooklikefrog [https://paste.ubuntu.com/p/gfTypZx3fk/ | 05:57 |
| T42 | <neochapay> Can som β¦], Build last droidmedia | 05:57 |
| T42 | <birdzhang> and build https://git.sailfishos.org/mer-core/nemo-gst-interfaces.git | 05:58 |
| T42 | <Pepelooklikefrog> Thanks | 06:05 |
| T42 | <Pepelooklikefrog> Now this: | 06:23 |
| T42 | <Pepelooklikefrog> https://paste.ubuntu.com/p/ms9KPYXt45/ | 06:23 |
| T42 | <edp_17> @elros34 : I wouldn't say we didn't do anything. π | 07:12 |
| T42 | <edp_17> I have amended bcm4330b1.conf, removed the masking of bluetooth-rfkill-event.service and rebooted with hoping it would work. | 07:13 |
| T42 | <edp_17> Not really. I manually need to kill brcm_patchram_plus every time before I am turning the bt on. | 07:14 |
| T42 | <edp_17> Plus, the device cannot pair with anything. | 07:14 |
| T42 | <edp_17> The changes I added into bcm4330b1.conf: | 07:15 |
| T42 | <edp_17> fork = 0 | 07:15 |
| T42 | <edp_17> #bdaddr = /efs/bluetooth/bt_addr - commented it out | 07:15 |
| T42 | <edp_17> no_flush = 0 - it was 1 before but I remember it was 0 before I started messing around | 07:16 |
| T42 | <edp_17> Now the brcm_patchram_plus starts automatically but it still has the --bd_addr 42π81:80:9a:4f parameter (I thought if I remove the bdaddr in config, that will remove this too) | 07:17 |
| r0kk3rz | why do you want to remove the mac addr? | 07:20 |
| T42 | <edp_17> Hi, because last night when I started the command manually it worked when I didn't include the mac addr parameter. | 07:21 |
| T42 | <edp_17> The system starts like this: 'brcm_patchram_plus --use_baudrate_for_download --no2bytes --enable_lpm --enable_hci --baudrate 3000000 --patchram /system/bin/bcm4330B1.hcd --bd_addr 42π81:80:9a:4f --scopcm 0,0,0,0,0,0,0,3,3,0 --tosleep 50000 /dev/ttySAC0' | 07:21 |
| T42 | <edp_17> but I cannot pair with anything. | 07:22 |
| T42 | <edp_17> If I start (manually) with this: '/usr/sbin/brcm_patchram_plus --enable_hci --enable_lpm --no2bytes --tosleep 50000 --baudrate 3000000 --use_baudrate_for_download --patchram /system/bin/bcm4330B1.hcd --scopcm=0,0,0,0,0,0,0,3,3,0 /dev/ttySAC0' | 07:22 |
| T42 | <edp_17> I can pair. | 07:22 |
| r0kk3rz | what mac address does the bt end up with? | 07:22 |
| T42 | <edp_17> When I run it manually? | 07:23 |
| r0kk3rz | both | 07:23 |
| T42 | <edp_17> I mean. When? if I run manually? | 07:23 |
| T42 | <edp_17> Sorry. | 07:23 |
| r0kk3rz | you will likely need to set it in the command, so maybe figure out why it doesnt work | 07:24 |
| T42 | <edp_17> I see. | 07:24 |
| T42 | <edp_17> How can I check the allocated mac address when I run manually? | 07:24 |
| r0kk3rz | i think its in the about screen in settings app | 07:25 |
| r0kk3rz | usually has various macs and imeis there | 07:25 |
| r0kk3rz | or theres bound to be some bluez command for it | 07:25 |
| T42 | <edp_17> Okay. I need to replicate what we done last night to see the mac address. Give me a few minutes. | 07:26 |
| r0kk3rz | without setting it, the mac addr will likely change every boot | 07:29 |
| T42 | <edp_17> Well, I have added back (uncommented) the bdaddr bit in the config and after the reboot, the device can be paired with others. But the bt only starts once with the device. If I switch it off cannot turn it on again. Even if I manually kill brcm_patchram_plus. | 07:31 |
| T42 | <edp_17> Now I go back to the setting we used last night. | 07:31 |
| T42 | <edp_17> Wait, you say I need to set the mac address to something persistent? | 07:32 |
| r0kk3rz | yeah | 07:32 |
| T42 | <edp_17> Now it got a new one when I killed the brcm_patchram_plus. | 07:33 |
| T42 | <edp_17> It was 9f:35:16:e6:ac:3e when the device started and it was 71πΏ9a:4d:31:87 after I killed the brcm_patchram_plus. | 07:34 |
| T42 | <edp_17> Now I replicated the settings we used last night. I masked the bluetooth-rfkill-event.service, rebooted and run the brcm_patchram_plus manually (without mac address) | 07:45 |
| T42 | <edp_17> BT turned on, it can be paired and the Bluetooth address is blank in About product. | 07:45 |
| T42 | <Verevka86> Please help | 10:10 |
| T42 | <Verevka86> error: external/vixl/Android.bp:130:1: "libvixl-arm64" depends on undefined module "dex2oat-pgo-defaults" | 10:10 |
| T42 | <Verevka86> ninja: build stopped: subcommand failed. | 10:10 |
| T42 | <Verevka86> ? | 10:46 |
| T42 | ζΈ©εͺη£ %lastname% was added by: ζΈ©εͺη£ %lastname% | 16:11 |
| dkter | huh, i actually get the same error about mkbootimg when i try to make hybris-hal again, even though it worked before | 18:32 |
| dkter | what i changed was i updated the droid-hal-device submodule | 18:32 |
| dkter | before it was on some commit from 2016. but it built fine then | 18:33 |
| dkter | actually no this is a slightly different error i think | 18:35 |
| dkter | ninja: error: 'system/core/mkbootimg/mkbootimg', needed by '/home/dkter/hadk/out/host/linux-x86/obj/EXECUTABL | 18:35 |
| dkter | ES/mkbootimg_intermediates/mkbootimg', missing and no known rule to make it | 18:35 |
| T42 | <elros34> 2016? So you didn't follow advice to create new droid-config/hal and you use few years old repos created for different android base. Expect more issues | 18:55 |
| dkter | i'm trying to update more things now | 18:59 |
| T42 | <C %lastname%> Hi guys, could anyone advice me about my Sailfish OS? I run 3.2.1.20 | 19:49 |
| T42 | <C %lastname%> on Sony Experia 10 | 19:49 |
| T42 | <C %lastname%> I think there was somewhere discussion about a bug and how to fix when Android apps dont lock in portraint as I need. | 19:50 |
| T42 | <C %lastname%> could anyone help, ideally DM me please | 19:50 |
| T42 | Pp %lastname% was added by: Pp %lastname% | 20:19 |
| T42 | <Pp %lastname%> Redmi note 4 | 20:19 |
| T42 | <C %lastname%> good news, its fixed in 3.3.0 version! | 20:20 |
| T42 | <C %lastname%> Androidβ’ App Support 8.1 | 20:21 |
| T42 | <C %lastname%> Updated to the latest android-8.1.0_r73 branch | 20:21 |
| T42 | <C %lastname%> Display orientation lock fixed for Android apps | 20:21 |
| T42 | <C %lastname%> WhatsApp: "add to contacts" and video playback fixed | 20:21 |
| T42 | <C %lastname%> anybody knows when it comes out please? | 20:21 |
| T42 | <kalube> For what phone? | 21:08 |
| T42 | <edp_17> @elros34 : Can I steel your time for the bt again? I think have managed to make a little progress. | 21:42 |
| T42 | <elros34> sure but I don't think I will be able to help you | 21:42 |
| T42 | <edp_17> Okay, maybe if I tell you what I have done and what the status now, will give you an idea. π | 21:43 |
| T42 | <edp_17> So, I masked bluetooth-rfkill-event.service and rebooted, like yesterday. | 21:43 |
| T42 | <edp_17> Then from /efs/imei/bt.txt I got a mac address for the bt. | 21:44 |
| T42 | <edp_17> Using this I amended the command I used yesterday to '/usr/sbin/brcm_patchram_plus --enable_hci --enable_lpm --no2bytes --tosleep 50000 --baudrate 3000000 --use_baudrate_for_download --patchram /system/bin/bcm4330B1.hcd --bd_addr 28:98:7B:24:40:8F --scopcm=0,0,0,0,0,0,0,3,3,0 /dev/ttySAC0' | 21:44 |
| T42 | <edp_17> This turned the bt on and the pairing with an other phone worked. | 21:45 |
| T42 | <elros34> so it was about correct mac? | 21:45 |
| T42 | <elros34> that is progress then | 21:46 |
| T42 | <edp_17> Pairing with my bt keyboard will work because it request me to enter the code but unfortunately on the screen I cannot see the code. | 21:46 |
| T42 | <elros34> tried 0000? | 21:46 |
| T42 | <edp_17> Yes. Didn't work. | 21:47 |
| T42 | <elros34> not sure if sfos even support such a pin prompt | 21:48 |
| T42 | <edp_17> I think that was the correct mac because the file name was bt.txt and inside it say: bt_macaddr: xyz | 21:48 |
| T42 | <edp_17> I think this is how it worked on my nexus 5 too. | 21:49 |
| T42 | <edp_17> I meant the pin prompt. | 21:49 |
| T42 | <elros34> so did you hardcode this mac in rfkill config file? Does it for if you add fork to options? | 21:50 |
| T42 | <elros34> work* | 21:51 |
| T42 | <edp_17> No, I didn't hardcode it yet. I have added the fork =0 into the config. (I don't know what this parameter for.) | 21:51 |
| T42 | <elros34> I mean did you tire to add fork when you run patchram manually | 21:52 |
| T42 | <elros34> s/tire/try/ | 21:52 |
| T42 | <edp_17> No, I left that out. | 21:53 |
| T42 | <edp_17> Now I hardcoded the mac address into config and removed masking. rebooting. | 21:53 |
| T42 | <edp_17> In the command the parameter is --bd_addr however in the config it is bdaddr=path_to_a_file. Shouldn't it be db_addr too in the config? | 21:55 |
| T42 | <edp_17> Well, it seems it doesn't use the hardcoded mac address. | 21:57 |
| T42 | <elros34> did you add bdaddr=/efs/imei/bt.txt? | 21:58 |
| T42 | <edp_17> No, I added bdaddr=28:98:7B:24:40:8F | 21:59 |
| T42 | <elros34> you just write "bdaddr=path_to_a_file" | 22:00 |
| T42 | <edp_17> Okay, I have changed and rebooting. | 22:00 |
| T42 | <edp_17> Is it a problem that the content of that file is: bt_macaddr:28987B24408F | 22:01 |
| T42 | <edp_17> So, not only a mac address but there are some text too. | 22:01 |
| T42 | <elros34> probably | 22:02 |
| T42 | <edp_17> Will see once it is started. | 22:02 |
| T42 | <elros34> or check sources: https://github.com/mer-hybris/bluetooth-rfkill-event/blob/master/bluetooth-rfkill-event/bluetooth_rfkill_event.c#L796. It expect 28:98:7B:24:40:8F format | 22:03 |
| T42 | <edp_17> Yep, in About product there is no mac address next to the bt, the 'ps ax | grep brcm' shows a different mac address what is in the /efs/imei/bt.txt file. | 22:03 |
| T42 | <edp_17> And the bt is not turning back on once I turned off. I need to stop the brcm_patchram_plus to be able to turn it on again. π | 22:05 |
| T42 | <edp_17> It does explain why I got different mac address every time when I kill brcm_patchram_plus and it restarts automatically. | 22:07 |
| T42 | <edp_17> So the question is where from should I get the correct bt mac address. | 22:07 |
| T42 | <elros34> how about just put it in some random file in correct format just for test | 22:08 |
| T42 | <edp_17> This is what I am thinking. What is the correct format? Only the mac address? With : as separator or only the characters? | 22:09 |
| T42 | <elros34> the one which has 17 chars | 22:10 |
| T42 | <edp_17> I have created a file /home/nemo/btmac with content of a 17 long mac address, I set this file in config at bdaddr, then killed brcm_patchram_plus, and now it doesn't restart automatically. | 22:12 |
| T42 | <edp_17> Even when I restart the bluetooth-rfkill-event.service, the brcm_patchram_plus doesn't start. | 22:14 |
| T42 | <edp_17> BUT, when I turned on the bt on device, it turned on!! | 22:15 |
| T42 | <edp_17> When I turned off and on again, it didn't do that. But when I killed the brcm_patchram_plus, I was able to turn it on again from settings. | 22:17 |
| T42 | <elros34> not sure what do you mean by it doesn't start. It should only start when you enable bluetooth in menu (rfkill is unblocked) and attach_hci() is called: https://github.com/mer-hybris/bluetooth-rfkill-event/blob/master/bluetooth-rfkill-event/bluetooth_rfkill_event.c#L1235 | 22:17 |
| T42 | <edp_17> So, it looks like that I need to kill brcm_patchram_plus every time when I turn the bt off, and it will work. | 22:17 |
| T42 | <elros34> try without fork=0 | 22:18 |
| T42 | <edp_17> I meant, it didn't turn on when I tap on the bluetooth menu in settings. | 22:18 |
| T42 | <edp_17> Oh, and the mac address in brcm_patchram_plus is the one I set in that file. | 22:19 |
| T42 | <elros34> ok, but based on your previous logs patchram should be killed automatically | 22:19 |
| T42 | <edp_17> No, it stays there. I need to kill it manually with kill -9. I don't know why. | 22:20 |
| T42 | <elros34> so have you tried to remove fork=0 from config? | 22:20 |
| T42 | <edp_17> Yes, I just did. It made it worse, because brcm_patchram_plus automatically restarts when I kill it and bt doesn't turn on when I tap on the bluetooth menu in settings. | 22:24 |
| T42 | <edp_17> Or I might need to restart the bluetooth-rfkill-event.service to take that fork change affect? | 22:25 |
| T42 | <elros34> obviously | 22:25 |
| T42 | <edp_17> Okay, start over. π | 22:26 |
| T42 | <edp_17> With fork=0 removed, bt turns on but after a second it turns off. | 22:28 |
| T42 | <elros34> rfkill-event logs with and without fork could be usefull | 22:34 |
| T42 | <elros34> a dirty hack which you couuld test is to kill patchram when rfkil change state in udev rules:) | 22:35 |
| T42 | <edp_17> Okay I'll do it. Can you show me an example for that, please? | 22:37 |
| T42 | <edp_17> is it a problem that I have two bluetooth type listed in rfkill? | 22:39 |
| T42 | <edp_17> 0 bluetooth bcm4330 Bluetooth unblocked unblocked | 22:39 |
| T42 | <edp_17> and | 22:39 |
| T42 | <edp_17> 13 bluetooth hci0 unblocked unblocked | 22:39 |
| T42 | <elros34> maybe, I have only one (with disabled and enabled bluetooth) | 22:41 |
| T42 | <edp_17> Which one do I need and how I can get rid of the other? | 22:42 |
| T42 | <elros34> I have no idea. About dirty hack: create some file in /lib/udev/rules.d/ https://pastebin.com/9rZ4rEUt | 22:44 |
| T42 | <edp_17> Thanks, I have added a file '100-bluetooth-brcm.rules' with that content. Do I need to restart something to make it work? | 22:49 |
| T42 | <elros34> I think: udevadm control --reload | 22:50 |
| T42 | <elros34> that second bluetooth in rfkil could explain your previous logs which I coudn't understand: "BT power driver unblocked" is called twice" and then free_hci is called | 22:52 |
| T42 | <elros34> or no, anyway full rfkill logs would be usefull to find out | 22:56 |
| T42 | <edp_17> I have tried the dirty hack and it seems I need to restart the bluetooth-rfkill-event.service when the rfkill status changes to 1. | 22:56 |
| T42 | <edp_17> Or not. | 22:57 |
| T42 | <edp_17> Now the bt turned on when I tap on bluetooth at the second time. | 22:58 |
| T42 | <edp_17> Yeah, it work like that. Here it is how it works (a bit odd): | 22:59 |
| T42 | <edp_17> starting status: bt is on | 22:59 |
| T42 | <edp_17> 1. tap on bluetooth in settings (turning bt off). Result brcm_patchram_plus is killed, bt is turned off, all paired listed devices disappear. | 23:00 |
| T42 | <edp_17> 2. come out of bluetooth menu and go back to the main menu in settings | 23:01 |
| T42 | <edp_17> 3. go back to bluetooth menu and tap on bluetooth (turning bt on). Result brcm_patchram_plus starts, but bt stays off, paired listed devices stays off. | 23:02 |
| T42 | <edp_17> 4. repeat step 2. | 23:02 |
| T42 | <edp_17> 5. repeat step 3, Result brcm_patchram_plus is killed, but stays off, paired listed devices stay off. | 23:03 |
| T42 | <edp_17> 6. repeat step 2. | 23:03 |
| T42 | <elros34> oh c'cmon :) | 23:03 |
| T42 | <edp_17> 7. repeat step 5, result bt turns on and listed devices appears. | 23:03 |
| T42 | <elros34> for me that is perfect:P | 23:04 |
| r0kk3rz | sounds good | 23:04 |
| T42 | <edp_17> I know, I know. But if I just tap on bluetooth without going back to main settings menu, it doesn't work. | 23:04 |
| T42 | <edp_17> It feels like if I stay in the bluetooth menu option, it only uses one action independently how many times I tap on the bluetooth option. I admit sounds very odd, but this is how it works. | 23:06 |
| T42 | <edp_17> Yeah, finally (with a little extra effort) the bt can be turned on/off. π | 23:07 |
| r0kk3rz | sounds like voodoo to me, its probably just taking time to turn on / off | 23:07 |
| T42 | <edp_17> Now I need to find out why the audio is not routed to the bt headset when it is connected. | 23:07 |
| T42 | <edp_17> @r0kk3rz: It can be. I try again without going out from the bluetooth menu. | 23:08 |
| T42 | <elros34> you should check how rfkill looks like, maybe killing patchram is not enough | 23:09 |
| T42 | <edp_17> Yes, you were right, it needs time and I don't need to go out from the bluetooth menu. But I still need to tap twice if I would like to turn it on. π | 23:11 |
| T42 | <edp_17> How do you mean 'how rfkill looks like'? | 23:11 |
| T42 | <edp_17> Plus I need to find a place for the hardcoded mac address somewhere else than home. π | 23:12 |
| T42 | <elros34> maybe you need to rfkill block hci or bcm4330*. Check whether you have mac saved in /data/property/ | 23:13 |
| T42 | <elros34> and please provide full verbose bluetoothrfkill-event log (without hack) someday | 23:14 |
| T42 | <edp_17> No, the mac address is not saved into there. I have two files in there persist.camera.shutter.disable and persist.media.metrics.enabled | 23:17 |
| T42 | <edp_17> How can I block hci or bcm4330? | 23:18 |
| T42 | <edp_17> How can I get bluetoothrfkill-event log? | 23:18 |
| T42 | <edp_17> journalctl? | 23:18 |
| T42 | <elros34> type rfkill block | 23:18 |
| T42 | <elros34> idx | 23:19 |
| T42 | <edp_17> I blocked bcm4330 and the bt turned off on the device. | 23:19 |
| T42 | <edp_17> The other hci somehow belongs to brcm_patchram_plus. Now the brcm_patchram_plus is not running, and rfkill has only 1 bluetooth (bcm4330). | 23:23 |
| Umeaboy | Will there be a solution to the mtd error after entering sfossdk? | 23:23 |
| Umeaboy | And number 2. The faq for the hadk doesn't mention if the patch process also will work on a local hybris-17.1 branch | 23:26 |
| Umeaboy | https://github.com/mer-hybris/hadk-faq/blob/master/README.rst#id21 | 23:27 |
| T42 | <edp_17> @elros34 : Here is the log for bt. (It was on, I turned off then turned on again. I needed to tap three time to get it turned on.) : https://paste.ubuntu.com/p/TwN5cJDRS7/ | 23:39 |
| r0kk3rz | the mtd error is benign | 23:40 |
| T42 | <elros34> @edp_17 that doesn't loooks like verbose log (DEBUG macros are not printed) and looks like hacky rule is used | 23:45 |
| T42 | <edp_17> Sorry, this is with the -o verbose : https://paste.ubuntu.com/p/pjvw3XTKjR/ | 23:48 |
| T42 | <edp_17> The hacky rule is what we have just done? | 23:48 |
| T42 | <edp_17> This log also includes that. | 23:48 |
| T42 | <elros34> I don't want to debug journal but rfkill-event: https://github.com/mer-hybris/bluetooth-rfkill-event/blob/master/bluetooth-rfkill-event/bluetooth_rfkill_event.c#L1114 | 23:50 |
| T42 | <elros34> or whatever option enable verbose mode in rfkill-event, I don't remember | 23:52 |
| T42 | <edp_17> Okay, I will try that. The hacky rule is how we finally make it work? | 23:54 |
| T42 | <edp_17> Do you know where to look for why the audio is not routed to the bt headset? | 23:57 |
| T42 | <elros34> how else you would call that udev rule? If you look at your loogs you can see that rfkill just crashh | 23:57 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!