you can get BT firmware from here 
https://github.com/wkennington/linux-firmware/tree/master/mediatek
thanks, i put the file to /lib/firmware, but still get the error
dmesg | grep hci0
[ 4.888999] bluetooth hci0: Direct firmware load for mediatek/mt7622pr2h.bin failed with error -2
[ 4.898991] Bluetooth: hci0: Failed to load firmware file (-2)
[ 5.058857] bluetooth hci0: Direct firmware load for mediatek/mt7622pr2h.bin failed with error -2
[ 5.068074] Bluetooth: hci0: Failed to load firmware file (-2)
root@bpi-iot-ros-ai:~# ls /lib/firmware/
ap6210 ap6212 brcm mt7622pr2h.bin
root@bpi-iot-ros-ai:~# bluetoothctl
[bluetooth]# power on
No default controller available
where does the driver look for the firmware? i have used kernel-documentation: https://www.kernel.org/doc/html/latest/driver-api/firmware/fw_search_path.html
oh i see that a subfolder mediatek is used…so i put it there…and after reboot i got this:
root@bpi-iot-ros-ai:~# ls /lib/firmware/mediatek
mt7622pr2h.bin
root@bpi-iot-ros-ai:~# dmesg | grep hci0
[ 14.819610] Bluetooth: hci0: Execution of wmt command timed out
[ 14.819672] Bluetooth: hci0: Failed to send wmt patch dwnld (-110)
first message seem to come from here: https://github.com/frank-w/BPI-R2-4.14/blob/4.19-r64-main/drivers/bluetooth/btmtkuart.c#L122 second from here: https://github.com/frank-w/BPI-R2-4.14/blob/4.19-r64-main/drivers/bluetooth/btmtkuart.c#L168
so mtk_setup_fw calls
mtk_hci_wmt_sync(hdev, MTK_WMT_PATCH_DWNLD, flag, dlen, fw_ptr);
and this calls
wait_on_bit_timeout(&bdev->tx_state, BTMTKUART_TX_WAIT_VND_EVT, TASK_INTERRUPTIBLE, HCI_INIT_TIMEOUT);
which failed with timeout, did i miss anything (btif should be enabled cause of removed disabled-line)
btif: serial@1100c000 {
so i looked here:
root@bpi-iot-ros-ai:~# cat /sys/firmware/devicetree/base/serial@1100c000/status
okay
root@bpi-iot-ros-ai:~# cat /sys/firmware/devicetree/base/serial@1100c000/bluetooth/status
okay
any idea about gpio? official site makes same calculation but with gpio22 which does not exist on 40pin-header (depending on shematic above)
@Jackzeng can you explain ATF-binary
ATF seems to stand for ARM trusted firmware…https://chromium.googlesource.com/external/github.com/ARM-software/arm-trusted-firmware/+/v0.4-rc2/docs/firmware-design.md
partitoninfo seems not actual because your image uses this partitiontable:
Disk /dev/sdb: 29.7 GiB, 31914983424 bytes, 62333952 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x674a82c6
Device Boot Start End Sectors Size Id Type
/dev/sdb1 204800 729087 524288 256M c W95 FAT32 (LBA)
/dev/sdb2 729088 14940159 14211072 6.8G 83 Linux
your script above says
# partition1 /dev/sdc1 vfat 204800~327679
# partition2 /dev/sdc2 ext4 327680~end
where can be the uboot-environment be saved (1MB on R2 in “user data area”)
how is the difference between emmc and SD (atf needed to)? same as on r2 (preloader to boot0, uboot to user-partition on same location without HEAD0/1)
how can we check from which device r64 has booted? (like we did in uboot for r2)
with bpi-r2 i can do
U-Boot> md.l 0x81DFFFF0 1
81dffff0: 434d4d65 eMMC
to see booted device…here emmc…if booted from sd-card i see this
BPI-R2> md.l 0x81DFFFF0 1
81dffff0: ffffffff ....
currently i don’t see emmc in uboot, but in linux i see mmc1blk1boot0 
so i can try to get emmc boot working
have now flashed mt-pack/mtk/bpi-r64/bin/preloader_bpi-r64_forsdcard-2k.img to mmcblk1boot0 and removed sd-card
F0: 102B 0000
F6: 3800 00A0
F3: 4000 0036
F5: 480A 0031
F5: 480A 0031
00: 1005 0000
F6: 3800 00A0
F3: 4000 0036
F5: 480A 0031
F5: 480A 0031
01: 102A 0001
02: 1005 0000
BP: 0000 00C0 [0001]
T0: 0000 0364 [000F]
System halt!
from sdcard this block looks like this:
F0: 102B 0000
F6: 3800 00A0
F3: 4000 0036
F5: 0000 0000
V0: 0000 0000 [0001]
00: 0000 0000
BP: 0000 0041 [0000]
G0: 0190 0000
T0: 0000 0448 [000F]
Jump to BL
is that right? i have not flashed uboot yet (need to compile it in ubuntu 4.14)…
flashed now also atf and uboot…
root@bpi-r64:~# mount /dev/mmcblk0p1 /boot
root@bpi-r64:~# ls /boot/
bananapi preloader_evb7622_64_forsdcard-2k.img
BPI-R2-HEAD1-512b.img System Volume Information
BPI-R2-HEAD440-0k.img u-boot-r64.bin
BPI-R64-atf.img u-boot-r64-fwu.bin
preloader_bpi-r64_forsdcard-2k.img
root@bpi-r64:~# dd of=/dev/mmcblk1 if=/boot/BPI-R64-atf.img bs=1k seek=512
256+0 records in
256+0 records out
262144 bytes (262 kB) copied, 0.0378364 s, 6.9 MB/s
root@bpi-r64:~# dd of=/dev/mmcblk1 if=/boot/u-boot-r64.bin bs=1k seek=768
391+1 records in
391+1 records out
400664 bytes (401 kB) copied, 0.0474521 s, 8.4 MB/s
but still “system halt”, does this mean that preloader is wrong or other bootstage? i also tried to change position of switch next to reset-button to 1…same result looking in log of sd-boot seems that preloader is not started…
T0: 0000 0454 [000F]
Jump to BL
UNIVPLL_CON0 = 0xFE000000!!!
mt_pll_init: Set pll frequency for 25M crystal
[PMIC_WRAP]wrap_init pass,the return value=0.
[pmic_init] Preloader Start..................
other preloader-file does also not work…or is preloader needed also on mmcblk1?
after compiling mmcutils for r64 and read Partitionconfig i see this:
root@bpi-r64:~# ./mmc extcsd read /dev/mmcblk1 | grep 'PARTITION_CONFIG'
Boot configuration bytes [PARTITION_CONFIG: 0x00]
so i tried to set it:
root@bpi-r64:~# ./mmc bootpart enable 1 1 /dev/mmcblk1
root@bpi-r64:~# ./mmc extcsd read /dev/mmcblk1 | grep 'PARTITION_CONFIG'
Boot configuration bytes [PARTITION_CONFIG: 0x48]
but boot is still not possible (also tried with bootswitch and reflashed preloader-bpi-r64-2k.img)…does preloader init emmc??
for sdcard i have it running except uboot…seems to need mtk-headers…with the u-boot-mtk.bin it works
can you explain what this do (what is LK)?
btw. i’ve managed to boot up a sd-card created from scratch (flash headers above, my modified uboot and boostrapped debian stretch) and also get netboot running on r64
uploaded image here:
https://drive.google.com/open?id=1rQCiUaR915BCqYG220DUm75bZQwt5N8i
繼續 Banana Pi BPI-R64 open source router with MTK MT7622 64 bit chip design 的討論:
There is a pending patch not in mainline kernel,
https://lkml.org/lkml/2018/12/13/29
could you help to apply the patch manually to your tree and see if it get betters?
if there is no the patch, the serial path won’t work properly.
please keep updates from https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/
all firmware for mediatek device would be kept at https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/mediatek
that is a separate tree apart from linux mainline kernel for firmware maintenance.
Had applied the patch allready applied
Used firmware from Banana Pi BPI-R64 open source router with MTK MT7622 64 bit chip design but it is not applied because of error above (same error with ryders/your patch
it looks like you also enabled spi0 if you use the default dts, try to disable spi0 at
spi0 already occupy the same pin 65 you want to software program in shell
https://elixir.bootlin.com/linux/v5.0-rc5/source/drivers/soc/mediatek/mtk-scpsys.c#L248
try to change the line from
usleep_range(12000, 12100);
to
msleep(50);
that is just a experiment to test whether power domain is not ready when bluetooth wants to be active
if issue is still, you can try below several times to see if it is always issue or seldom issue.
hciconfig hci0 down
hciconfig hci0 up
hciconfig hci0 down
hciconfig hci0 up
…
1 Like
How do you get spi0-relationship?
I try gpio56=pin3 spi0 is pin 31,32,36 according to shematic (= gpios 82,64,65)
pin 56 might be occupied by i2c1 if you used the default dts from upstream
you should disable the node before you want to software program that pin
1 Like
gpio-issue solved so far…gpio was blocked by activated i2c1…so if the special-function is enabled in dts, the direct gpio-access is blocked
root@bpi-r64:~# GPIO_NO=$((409+56))
root@bpi-r64:~# echo $GPIO_NO
465
root@bpi-r64:~# echo $GPIO_NO > /sys/class/gpio/export
root@bpi-r64:~# echo out > /sys/class/gpio/gpio${GPIO_NO}/direction
root@bpi-r64:~# echo 1 > /sys/class/gpio/gpio${GPIO_NO}/value
root@bpi-r64:~# echo 0 > /sys/class/gpio/gpio${GPIO_NO}/value
my led goes on and off
root@bpi-r64:~# apt-get install bluez
...
root@bpi-r64:~# uname -a
Linux bpi-r64 4.19.0-r64-main #70 SMP PREEMPT Sat Feb 9 09:53:04 CET 2019 aarch64 GNU/Linux
root@bpi-r64:~# dmesg | grep hci0
root@bpi-r64:~#
root@bpi-r64:~# bluetoothctl
[NEW] Controller 00:00:46:76:22:01 bpi-r64 [default]
[bluetooth]# power on
Changing power on succeeded
[bluetooth]# scan on
Discovery started
[CHG] Controller 00:00:46:76:22:01 Discovering: yes
[bluetooth]# [ 134.372291] Bluetooth: hci0: last event is not cmd complete (0x0f)
...
[CHG] Device B4:9D:0B:5F:2A:B9 Name: Aquaris X5 Plus
...
[CHG] Device B4:9D:0B:5F:2A:B9 ServicesResolved: yes
[CHG] Device B4:9D:0B:5F:2A:B9 Paired: yes
Pairing successful
[CHG] Device B4:9D:0B:5F:2A:B9 ServicesResolved: no
[CHG] Device B4:9D:0B:5F:2A:B9 Connected: no
[bluetooth]# [ 207.332516] Bluetooth: hci0: last event is not cmd complete (0x0f)
[bluetooth]# exit
[DEL] Controller 00:00:46:76:22:01 bpi-r64 [default]
root@bpi-r64:~#
except of the event-info/warning all is good now
Bluetooth: hci0: last event is not cmd complete (0x0f)
the change seems to create problems detecting mmc-rootfs
sometimes i get this message and boot-prcess hangs
Waiting for root device /dev/mmcblk0p2...
@linkerosa have you an idea regarding preloader and emmc?
Hello, I want to attach an SSD and need SATA 3 speed (ie. 6 Gbps). Which of the boards has all these features: Gigabit Ethernet, USB 3, SATA 3 ? Thx
Update: in the following thread someone wrote “The BPI-R2 has two SATA 3.0 Ports / 6 Gbps and not SATA II” ( SATA Performance + Max Disk Size ). But in many product descriptions only SATA 2 is mentioned, also in the Hardware spec wiki ( http://wiki.banana-pi.org/Banana_Pi_BPI-R2#Documents ) it says “Two SATA 2.0 Port”
What is correct?
Regardless if it is sata 3,you will not reach the speed of 6gbit/s. Why does it matter for you?
Btw. This thread is for r64
Hi, I’m just a newbie, don’t know yet the features of the different boards, I wongly assumed the R64 is the successor to R2; it seems it is rather the W2.
I just wanted evaluate the possibilities of building a backup-box to be used in LAN with multiple client-machines to backup, so the LAN and the storing to the SSD should be as fast as possible as this shall be a “realtime” backup solution, ie. always on, though incremental, and it should serve as many clients as possible. So I assumed getting full SATA III speed, but later learned that it’s not a full SATA III like you remarked too. I need to find a cheap HW for this project. Today I found another one (Rock Pi 4 for $39), it looks promising, but it lacks SATA; has only USB3, and not clear yet if one can use Debian on it. It’s not easy to find the right HW for the project as I’m more a SW guy…
imho most SOCs don’t have native SATA-support…i don’t know how it realized on R2/R64, imho as pcie-sata-bridge, where mini-pcie has limitation to 500Mbit/s, so it is not possible to reach sata2/3-throughput…but please make a new thread if needed
for GPIO, how to disable pwm-mapping?
i get this error for pin7 (gpio101):
root@bpi-r64:~# GPIO_NO=$((409+101))
root@bpi-r64:~# echo $GPIO_NO > /sys/class/gpio/export
[ 365.604377] mtk-pinctrl 10211000.pinctrl: pin PWM7 already requested by 11006000.pwm; cannot claim for pinctrl_mt7622:510
[ 365.615703] mtk-pinctrl 10211000.pinctrl: pin-101 (pinctrl_mt7622:510) status -22
-bash: echo: write error: Invalid argument
but nice that here is a message for blocked access
can i disable it from userspace or only in dts?
it seems it is this section in dts:
only pwm7 seems to be defined, why not the others? maybe we can define pwm1-7 (except 3, i’ve did not found on header)
can i use generic pwm-node and define subnodes for pwm1-7?
something like this (of course i have to define the pwmx_pins):
&pwm {
pinctrl-names = "default";
status = "okay";
pwm1{
pinctrl-0 = <&pwm1_pins>;
}
pwm2{
pinctrl-0 = <&pwm2_pins>;
}
pwm4{
pinctrl-0 = <&pwm4_pins>;
}
...
}maybe we can define pwm1-7 (except 3, i’ve did not found on header)
Sure! you could add it if you want to do that.
can i use generic pwm-node and define subnodes for pwm1-7? something like this (of course i have to define the pwmx_pins):
You can take a look at i2c/spi nodes to understand how to add “generic pin configuration” (via function/group) in dts
how can i use emmc? it seems the sd-card-preloader does not work for emmc, atf should be loaded after uboot, but i don’t get into uboot
btw. have not get pwm working yet…also no answer from sean
BPI-R64 MT7622 schematic diagram public
1 Like