Package Details: pi-bluetooth 1.2_4-1

Git Clone URL: (read-only)
Package Base: pi-bluetooth
Description: Services, firmware and udev rules to get integrated bluetooth running in Raspberry Pi 3
Upstream URL: None
Licenses: Multiple
Submitter: Huulivoide
Maintainer: blm768
Last Packager: blm768
Votes: 24
Popularity: 0.148833
First Submitted: 2016-03-19 20:54
Last Updated: 2019-04-12 03:51

Latest Comments

1 2 3 4 Next › Last »

keithspg commented on 2019-04-27 15:31

Still struggling here. I do not know how much of my problem is this and how much is with bluealsa. With a BT dongle I have no problems with bluealsa. I pair and connect and it does not seem to matter what phone I use. It seems to always connect and play. With the onboard modules, I get nothing but problems. The one I have had the most luck with is the B3+. It seems to play music with any phone I pair and seems to be robust. The B3 is a nightmare and the Zero I have had no luck getting any output at all recently (still working on figuring out why). On teh B3, I can pair and connect, eventually, and send audio, but it will disconnect and with certain devices (samsung Tab4), I get choppy audio. I get smooth audio with Moto Z2Play and HTC M9. On the B3+ all seem to work with smooth audio.

The B3+ works with ExecStart=/usr/bin/hciattach -n /dev/ttyAMA0 bcm43xx 3000000 flow -

The B3 I am using: ExecStart=/usr/bin/hciattach -n /dev/ttyAMA0 bcm43xx 921600 noflow -

The B3+ device is at hci1 (do not really know why) and the b3 device is hci0. On thh B3+, with the btattach command, I was able to blacklist a module and get the device to show up at hci0. That does not work with the hciattach command. I also tried the AUR alarm-bluetooth-raspberrypi and noted that the behavior on audio was identical to the btattach command (choppy audio with all paired devices), but the dmesg actually showed what was going on. With this command, the response is different in dmesg (no hci device is announced). On the B3+:

dmesg | grep hci

[ 4.298943] sdhci: Secure Digital Host Controller Interface driver [ 4.298945] sdhci: Copyright(c) Pierre Ossman [ 4.299395] sdhci-pltfm: SDHCI platform and OF driver helper

dmesg | grep Blue

[ 7.498883] Bluetooth: Core ver 2.22 [ 7.498933] Bluetooth: HCI device and connection manager initialized [ 7.498946] Bluetooth: HCI socket layer initialized [ 7.498952] Bluetooth: L2CAP socket layer initialized [ 7.498973] Bluetooth: SCO socket layer initialized [ 7.502178] Bluetooth: Generic Bluetooth SDIO driver ver 0.1 [ 8.814895] Bluetooth: HCI UART driver ver 2.3 [ 8.814916] Bluetooth: HCI UART protocol H4 registered [ 8.814924] Bluetooth: HCI UART protocol BCSP registered [ 8.814979] Bluetooth: HCI UART protocol LL registered [ 8.814982] Bluetooth: HCI UART protocol ATH3K registered [ 8.815007] Bluetooth: HCI UART protocol Three-wire (H5) registered [ 8.815198] Bluetooth: HCI UART protocol Broadcom registered [ 9.012920] Bluetooth: BNEP (Ethernet Emulation) ver 1.3 [ 9.012928] Bluetooth: BNEP filters: protocol multicast [ 9.012944] Bluetooth: BNEP socket layer initialized

blm768 commented on 2019-04-24 03:12

I haven't done any testing on AArch64. However, based on the GitHub issue comments (, AArch64 support for some of the devices appears to be rather incomplete, and it looks like the Bluetooth controller may be one of those devices.

bipbip1981 commented on 2019-04-23 19:10

Hi, On my RPI3B with archlinux ARM AARCH64, i have two rfollowing results with dmesg and systemd

On my RPI3B with archlinux ARM armor, all it good

Bluetooth not work a this time with AARCH64? i do a mistake somewhere? i miss explain somewhere in documentation?

Thank a lot for helping

Sorry for my poor english

systemctl status brcm43438

  • brcm43438.service - Broadcom BCM43438 bluetooth HCI Loaded: loaded (/usr/lib/systemd/system/brcm43438.service; enabled; vendor preset: disabled) Active: inactive (dead) Condition: start condition failed at Tue 2019-04-23 21:00:23 CEST; 6s ago `- ConditionPathIsDirectory=/proc/device-tree/soc/gpio@7e200000/bt_pins was not met

Apr 23 20:53:41 raspberrypi systemd[1]: Condition check resulted in Broadcom BCM43438 bluetooth HCI being skipped.


[ 31.194545] Bluetooth: hci0: BCM: failed to write update baudrate (-110) [ 31.207605] Bluetooth: hci0: Failed to set baudrate [ 31.338676] Bluetooth: hci0: BCM: chip id 94 [ 31.351179] Bluetooth: hci0: BCM: features 0x2e [ 31.384607] Bluetooth: hci0: BCM43430A1 [ 31.394081] Bluetooth: hci0: BCM43430A1 (001.002.009) build 0000

blm768 commented on 2019-04-16 04:19

Based on my research, the 3B may require a different baud rate with control flow turned off. Does replacing the hciattach command with "/usr/bin/hciattach /dev/ttyAMA0 bcm43xx 921600 noflow -" work on your 3B?


keithspg commented on 2019-04-15 02:23

Just tried this on a Pi3b+ and it works if I set the bluealsa device as hci1. The problem is that this is the default device and it shows up on hci1. How do I get the default device to be hci0.

From btmgmt, it shows this: [mgmt]# info Index list with 1 item hci1: Primary controller addr 43:45:C0:00:1F:AC version 8 manufacturer 15 class 0x040000 supported settings: powered connectable fast-connectable discoverable bondable link-security ssp br/edr hs le advertising secure-conn debug-keys privacy static-addr current settings: powered connectable bondable ssp br/edr le secure-conn name gusrune short name

it does not list an hci0, but when it enumerates on the B3+, it shows up as hci1. All this on a PiB3+

On another pi, a regular B3 (not a plus) I have Bluealsa set up identically installed the same versions of pi-bluetooth, etc. WHen I run brcm43438, the device shows up as hci0

[mgmt]# info Index list with 1 item hci0: Primary controller addr B8:27:EB:D7:82:70 version 7 manufacturer 15 class 0x040000 supported settings: powered connectable fast-connectable discoverable bondable link-security ssp br/edr hs le advertising secure-conn debug-keys privacy static-addr current settings: powered connectable discoverable bondable ssp br/edr le secure-conn name gretarune short name

after trying numerous times and finally getting it to pair and connect, when it plays, the timing is way off (fast) and it cuts in and out. This is only on the B3. B3+ is fine with playback. I have not yet tried on a Zero. Any ideas?

keithspg commented on 2019-04-14 19:26

I just tried this latest version. Uninstalled my hack and installed this one. Only had opportunity to try it on a Pi3 so far, but no joy. I can pair, but bluealsa cannot find the 'adapter'

bluealsa[2255]: /usr/bin/bluealsa: Adapter not available: /org/bluez/hci0/dev_BC_FF_EB_39_E0_5E

rebooted 2x. no help

blm768 commented on 2019-04-12 03:57

I've pushed a new version of the package that seems to work properly on my Pi; for whatever reason, the old hciattach utility seems to be more stable on this hardware. Using that, the updated *-sdio.txt files don't seem to be necessary, at least according to my tests. Thanks, @keithspg; your research was super useful.

keithspg commented on 2019-04-10 14:27

@blm768 Thanks for looking into this. I made a PKGBUILD and .install file for my use to get the files in the correct places. Apparently hciattach needs the HCD files in /etc/firmware for it to work. A link to my files is here: I have been working on getting this set up so my Pi is a BT enabled speaker (a2dp-sink) with bluez-alsa. I have success with a B3+ and a B3. It would also be helpful if we could understand why the a2db device will work only on hci1 on a B3+ and it shows up at hci0 (where I expect it) on a B3. I have not yet been able to get the Zero going yet.

blm768 commented on 2019-04-10 05:12

@keithspg Well, the utilities in bluez-utils-compat are considered deprecated, but if they work, they work. I'm able to reproduce the choppy audio issue on my own device. Seems like changing the btattach parameters to what should be equivalent to the old hciattach command (/usr/bin/btattach -B /dev/ttyAMA0 -P bcm -S 921600 -N) doesn't fix the issue, either. I might have to experiment with the compat package again.

keithspg commented on 2019-04-10 01:54

I finally got this working. I hope someone here can help tell me if I am doing it right. This list had all the 'choppy audio' comments in it. I was able to get this working using pieces of an older commit from this AUR. Plus I grabbed the edits to the txt files as well. The nvram txt files: