ah yea but those are inherent to the PI | adafruit | Page 1

strange ridge Jul 26, 2023, 4:02 PM

#

we're too chatty 🙂 ok here is a thread

serene lichen Jul 26, 2023, 4:02 PM

#

🙂

strange ridge Jul 26, 2023, 4:03 PM

#

what am i lookin for in dmesg

serene lichen Jul 26, 2023, 4:03 PM

#

ah and maybe something went wrong with the drm flag for cmdline, you should see a lot of verboose mesages like this

[  184.619666] [drm:drm_stub_open [drm]]
[  184.619785] [drm:drm_open [drm]] comm="Xorg", pid=5384, minor=0
[  184.619887] [drm:drm_ioctl [drm]] comm="Xorg" pid=5384, dev=0xe200, auth=1, DRM_IOCTL_MODE_GETRESOURCES
[  184.619972] [drm:drm_ioctl [drm]] comm="Xorg", pid=5384, ret=-95
[  184.620060] [drm:drm_release [drm]] open_count = 1
[  184.620142] [drm:drm_file_free.part.0 [drm]] comm="Xorg", pid=5384, dev=0xe200, open_count=1
[  184.620247] [drm:drm_release [drm]]
[  184.620329] [drm:drm_release [drm]] driver lastclose completed
[  184.620477] [drm:drm_stub_open [drm]]
[  184.620567] [drm:drm_open [drm]] comm="Xorg", pid=5384, minor=0
[  184.620662] [drm:drm_ioctl [drm]] comm="Xorg" pid=5384, dev=0xe200, auth=1, DRM_IOCTL_MODE_GETRESOURCES
[  184.620747] [drm:drm_ioctl [drm]] comm="Xorg", pid=5384, ret=-95
[  184.620833] [drm:drm_release [drm]] open_count = 1
[  184.620929] [drm:drm_file_free.part.0 [drm]] comm="Xorg", pid=5384, dev=0xe200, open_count=1
[  184.621016] [drm:drm_release [drm]]
[  184.621096] [drm:drm_release [drm]] driver lastclose completed
[  184.622168] [drm:drm_stub_open [drm]]
[  184.622291] [drm:drm_open [drm]] comm="Xorg", pid=5384, minor=0
[  184.622394] [drm:drm_ioctl [drm]] comm="Xorg" pid=5384, dev=0xe200, auth=1, DRM_IOCTL_MODE_GETRESOURCES
[  184.622480] [drm:drm_ioctl [drm]] comm="Xorg", pid=5384, ret=-95
[  184.622569] [drm:drm_release [drm]] open_count = 1
[  184.622651] [drm:drm_file_free.part.0 [drm]] comm="Xorg", pid=5384, dev=0xe200, open_count=1
[  184.622741] [drm:drm_release [drm]]
[  184.622820] [drm:drm_release [drm]] driver lastclose completed

strange ridge Jul 26, 2023, 4:03 PM

#

root@devpi:/home/pi# dmesg | grep drm
[ 0.000000] Kernel command line: coherent_pool=1M 8250.nr_uarts=1 snd_bcm2835.enable_headphones=0 snd_bcm2835.enable_headphones=1 snd_bcm2835.enable_hdmi=1 bcm2708_fb.fbwidth=720 bcm2708_fb.fbheight=480 bcm2708_fb.fbswap=1 vc_mem.mem_base=0x3ec00000 vc_mem.mem_size=0x40000000 console=ttyS0,115200 console=tty1 root=PARTUUID=99c577d8-02 rootfstype=ext4 fsck.repair=yes rootwait loglevel=6 drm.debug=0xf
[ 5.992187] systemd[1]: Starting Load Kernel Module drm...
[ 6.297088] [drm:drm_core_init [drm]] Initialized
[ 6.301348] systemd[1]: modprobe@drm.service: Succeeded.
[ 6.302452] systemd[1]: Finished Load Kernel Module drm.

#

🤔

serene lichen Jul 26, 2023, 4:07 PM

#

hum, I am running a year old system here so maybe that changed in more recent kernels
but that alone doesn't seem to give me the log messages for DSI either currently. Lemme check my notes if there was more to it to enable those

strange ridge Jul 26, 2023, 4:07 PM

#

ok do you mind burning a new pi sd with raspbian lite

serene lichen Jul 26, 2023, 4:07 PM

#

but the icn dts doesn't show anything either?

#

yea lemme do that realy quick

strange ridge Jul 26, 2023, 4:08 PM

#

ok me too ill start fresh since pi's quickly get crusty with wierdo setups 😄

#

serene lichen Jul 26, 2023, 4:09 PM

#

ah lets maybe use a full distro, I'm not sure the lite still includes the desktop at all?

#

that could maybe explain why the drm log doesn't post anything, not entirely sure

#

I run headless here too but it had the desktop installed so I think that was a full sized Raspberry PI OS

#

serene lichen Jul 26, 2023, 4:13 PM

#

serene lichen that could maybe explain why the drm log doesn't post anything, not entirely sur...

my thinking is that they do not start up drm at all there and without it I think you are missing the kms driver as well to output dsi

strange ridge Jul 26, 2023, 4:20 PM

#

sorry had a baby thing to check

#

oh yeah gooooood point, i do recall that tinydrm you have to 'kick' it by sending data to the framebuf

serene lichen Jul 26, 2023, 4:20 PM

#

no worries! currently booting up the fresh system

strange ridge Jul 26, 2023, 4:20 PM

#

ok let me reburn 🙂

#

#

ok burned, do you put an hdmi display on?

serene lichen Jul 26, 2023, 4:27 PM

#

no just Ethernet and the DSI cable

strange ridge Jul 26, 2023, 4:28 PM

#

ok let me boot up and ssh in

serene lichen Jul 26, 2023, 4:29 PM

#

ah but I failed to setup stuff through the imager, if you haven't either you might get stuck where I am with finishing the first boot setup because the default pw thing is no more

strange ridge Jul 26, 2023, 4:29 PM

#

uhhhh i think i found my problem. its a pi3 not pi 4 😭

serene lichen Jul 26, 2023, 4:29 PM

#

oh! hm but I'm not sure if it should make a huge difference

strange ridge Jul 26, 2023, 4:30 PM

#

yeah but i was putting the dto stuff under the [pi4] section hah

#

ok np

serene lichen Jul 26, 2023, 4:30 PM

#

ah right was just checking that haha

strange ridge Jul 26, 2023, 4:30 PM

#

i think i changed my imager defaults to have a password

#

guess ill find out

serene lichen Jul 26, 2023, 4:30 PM

#

useful they added that but easy to miss too

strange ridge Jul 26, 2023, 4:31 PM

#

sigh it erased my default settings, ok reburning

#

at least its fast on USB 3 + class 10

serene lichen Jul 26, 2023, 4:34 PM

#

yea those newer UHS V30/60 cards are really quick!
flashes the image in around 40 seconds

strange ridge Jul 26, 2023, 4:35 PM

#

ooh that sounds even better

#

maybe ill get some of those

#

ok done

#

ok i'm in!

#

send me what you're running and ill duplicate

serene lichen Jul 26, 2023, 4:39 PM

#

just finished setup, one sec

#

ok getting an error atm for dsi

#

but also found the flag that enabled device tree debug messages
so in config.txt I have added to [all]

dtoverlay=i2c0,pins_44_45
dtoverlay=ICN6211
dtdebug=on

the i2c only needed if you use the I2C over the DSI cable

#

which atm leads
[ 6.850621] vc4_dsi fe700000.dsi: Failed to create device link (0x180) with 1.panel_disp1

strange ridge Jul 26, 2023, 4:51 PM

#

ok wanna keep poking at it to see if you can fix?

#

this is tragically common with raspi, they like to keep up to date which....means a lot of stuff is constantly breaking 🙃

serene lichen Jul 26, 2023, 4:53 PM

#

yea 😦
sorry for the wild goose hunt, should have tested with a new image sooner

#

it means the device tree file is loading at least but sounds like something in their DT setup changed

#

some of the warnings you got in the compilation (and I also now got) were not all the typical stuff, some of that is definitely new warnings

strange ridge Jul 26, 2023, 4:56 PM

#

@serene lichen ok im gonna be in another window. if you dont wanna work on this now, totally cool just send an email if/when ready for a second test

#

you may have some luck decompiling the 7 inch dtbo

serene lichen Jul 26, 2023, 5:17 PM

#

ok got it

#

dumb mistake caused by previous outdated assumptions

strange ridge Jul 26, 2023, 5:18 PM

#

not dumb, think of how much we are learning 🙂

serene lichen Jul 26, 2023, 5:18 PM

#

you re absoultey right, still trying to adjust my language with these things 🙂

#

so the reason the offical 7inch overlay doesn't work is because it does a i2c probe for their MCU, that fails and it doesn't proceed to enable DSI. Previously I had worked with the ICN via one of those clone display that emulate that IC while attaching my own panels to their breakout, that then worked just fine with the standard overlay because the only difference between those bridge IC drivers is their IC configuration, the DSI aspect is the same and we don't really care about the config here.

In my overlay I had the actual ICN driver loaded for the bridge fragment, the Pi OS image by default does not include this driver and that seems to fail rather silently.
tldr; fix: need to change to line 26 to
compatible = "toshiba,tc358762"; in the dts

#

strange ridge Jul 26, 2023, 5:28 PM

#

hi back was getting some 🍵

#

ok let me scp that over

#

yah i do not want to 'mimic' their i2c

#

i know DF robot does, but i think its tacky

#

ok did you isntall the blinka script to get the ICN i2c going

serene lichen Jul 26, 2023, 5:30 PM

#

its the "lazy" route for sure and only really works usefully if you use the same resolution, if you don't will have to touch kernel drivers anyways. What it gets you though is bootloader level DSI as currently only their panel is supported in the firmware

serene lichen Jul 26, 2023, 5:31 PM

#

strange ridge ok did you isntall the blinka script to get the ICN i2c going

I did but actually not sure if I re-ran it on the new system for the config

strange ridge Jul 26, 2023, 5:31 PM

#

oh so with the 7" you will get the rainbow test-square

#

but with this setup its like, whenever the dsi dtb is loaded after a few secs?

serene lichen Jul 26, 2023, 5:31 PM

#

yea

#

and the bootloader menu is you boot from a different device

strange ridge Jul 26, 2023, 5:31 PM

#

i think thats ok

#

theres a bootloader menu!??

serene lichen Jul 26, 2023, 5:32 PM

#

strange ridge but with this setup its like, whenever the dsi dtb is loaded after a few secs?

yea essentially as soon as the kernel becomes alive

serene lichen Jul 26, 2023, 5:32 PM

#

strange ridge theres a bootloader menu!??

fairly new yea! It gives you lots of debug info too, quite handy

#

well not a real menu I guess but it tells you whats happening and what its trying to do which is a bit more handy than the old blinky error codes

strange ridge Jul 26, 2023, 5:33 PM

#

being headless has me missing out

serene lichen Jul 26, 2023, 5:33 PM

#

transfering config on the new image also seems to work fine

strange ridge Jul 26, 2023, 5:33 PM

#

ok im configuring this setup ...

#

sudo update-alternatives --install /usr/bin/python python /usr/bin/python2.7 1
sudo update-alternatives --install /usr/bin/python python /usr/bin/python3.7 2
#update-alternatives --config python
python --version
pip3 install RPi.GPIO
sudo pip3 install --upgrade setuptools
cd ~
sudo pip3 install --upgrade adafruit-python-shell
wget https://raw.githubusercontent.com/adafruit/Raspberry-Pi-Installer-Scripts/master/raspi-blinka.py
sudo python3 raspi-blinka.py

serene lichen Jul 26, 2023, 5:34 PM

#

ah! handy command if you want to verify the overlay loaded ok: sudo vcdbg log msg

strange ridge Jul 26, 2023, 5:34 PM

#

ok hold on i have to do an apt update and all that jazz

#

this might be like 5 minutes - ill post when ready ⏲️

serene lichen Jul 26, 2023, 5:37 PM

#

ok 🙂

strange ridge Jul 26, 2023, 5:48 PM

#

ok successfully installed blinka, wrote config to icn

#

now let me compile the dts

#

still got many warns

#

sudo cp ICN6211.dtbo /boot/overlays/
ICN6211.dts:58.5-15: Warning (reg_format): /fragment@1/__overlay__/panel_disp1@0:reg: property has invalid length (4 bytes) (#address-cells == 2, #size-cells == 1)
ICN6211.dts:71.5-15: Warning (reg_format): /fragment@1/__overlay__/reg_bridge@0:reg: property has invalid length (4 bytes) (#address-cells == 2, #size-cells == 1)
ICN6211.dtbo: Warning (pci_device_reg): Failed prerequisite 'reg_format'
ICN6211.dtbo: Warning (pci_device_bus_num): Failed prerequisite 'reg_format'
ICN6211.dtbo: Warning (simple_bus_reg): Failed prerequisite 'reg_format'
ICN6211.dtbo: Warning (i2c_bus_reg): Failed prerequisite 'reg_format'
ICN6211.dtbo: Warning (spi_bus_reg): Failed prerequisite 'reg_format'
ICN6211.dts:57.31-68.6: Warning (avoid_default_addr_size): /fragment@1/__overlay__/panel_disp1@0: Relying on default #address-cells value
ICN6211.dts:57.31-68.6: Warning (avoid_default_addr_size): /fragment@1/__overlay__/panel_disp1@0: Relying on default #size-cells value
ICN6211.dts:70.29-77.6: Warning (avoid_default_addr_size): /fragment@1/__overlay__/reg_bridge@0: Relying on default #address-cells value
ICN6211.dts:70.29-77.6: Warning (avoid_default_addr_size): /fragment@1/__overlay__/reg_bridge@0: Relying on default #size-cells value
ICN6211.dtbo: Warning (avoid_unnecessary_addr_size): Failed prerequisite 'avoid_default_addr_size'
ICN6211.dtbo: Warning (unique_unit_address): Failed prerequisite 'avoid_default_addr_size'

#

alright so im adding the dtoverlay to /boot/config

#

dtoverlay=ICN6211

#

should i still do the cmdline debug step too?

#

pi@devpiX:~ $ dmesg | grep dsi
[ 0.051794] platform 3f101000.cprman: Fixed dependency cycle(s) with /soc/dsi@7e700000
[ 9.300934] vc4_dsi 3f700000.dsi: Failed to create device link (0x180) with 0.panel_disp1
[ 10.869572] vc4-drm soc:gpu: bound 3f700000.dsi (ops vc4_dsi_ops [vc4])
pi@devpiX:~ $

#


[pi4]
# Run as fast as firmware / board allows
arm_boost=1

[all]
enable_uart=1
#dtoverlay=i2c0,pins_44_45
dtoverlay=ICN6211
dtdebug=on

#

[    9.300934] vc4_dsi 3f700000.dsi: Failed to create device link (0x180) with 0.panel_disp1

serene lichen Jul 26, 2023, 5:57 PM

#

strange ridge still got many warns

yea had them too, haven't looked into them yet but don't seem criticial

strange ridge Jul 26, 2023, 5:57 PM

#

pi@devpiX:~ $ sudo vcdbg log msg
Failed to allocate -201253077 bytes for message buffer

#

????

serene lichen Jul 26, 2023, 5:57 PM

#

oh weird

strange ridge Jul 26, 2023, 5:58 PM

#

lol

#

should i wrangle a pi 4 - maybe something does make a diff

serene lichen Jul 26, 2023, 5:58 PM

#

loglevel=6 drm.debug=0x14is what I currently have appended in my cmdline.txt but that won't fix that vcdbg msg

serene lichen Jul 26, 2023, 5:59 PM

#

strange ridge should i wrangle a pi 4 - maybe something does make a diff

yea can't hurt to remove more differences

strange ridge Jul 26, 2023, 5:59 PM

#

console=serial0,115200 console=tty1 root=PARTUUID=bd563b1b-02 rootfstype=ext4 fsck.repair=yes rootwait quiet splash plymouth.ignore-serial-consoles loglevel=6 drm.debug=0x14

#

let me reboot

#

kinda rad

#

luv when so much error log it wraps

serene lichen Jul 26, 2023, 5:59 PM

#

haha

strange ridge Jul 26, 2023, 6:00 PM

#

ok i do have a pi4

#

[    9.726703] vc4_dsi 3f700000.dsi: Failed to create device link (0x180) with 0.panel_disp1

#

still

#

oh wow lots of error messages

#

one sec

#

📎 message.txt

serene lichen Jul 26, 2023, 6:02 PM

#

oh funnily enough I still get that error too but its outputting DSI happily lol

strange ridge Jul 26, 2023, 6:02 PM

#

huge

#

huh sorta is implying that it is outputting 800x480

#

let me see if theres any signal on the dsi lines?

#

no nothing 🤔

#

ok what if i put this onto my pi4

serene lichen Jul 26, 2023, 6:04 PM

#

strange ridge

yea that looks like a similar setup log as I'm getting atm

strange ridge Jul 26, 2023, 6:04 PM

#

the DSI cable is A-B type right?

#

contacts on opposite sides at the ends

serene lichen Jul 26, 2023, 6:05 PM

#

yes

strange ridge Jul 26, 2023, 6:05 PM

#

(worth checking)

serene lichen Jul 26, 2023, 6:06 PM

#

serene lichen Jul 26, 2023, 6:06 PM

#

strange ridge (worth checking)

for sure

serene lichen Jul 26, 2023, 6:07 PM

#

strange ridge ``` pi@devpiX:~ $ sudo vcdbg log msg Failed to allocate -201253077 bytes for mes...

googling that it seems to pop up for a lot of people on Pi3's
https://github.com/raspberrypi/firmware/issues/1670

strange ridge Jul 26, 2023, 6:07 PM

#

lol

#

welp my pi 4 isnt booting and im somewhat running out of time today - i have a hackchat at 2pm and then 2 hours of video at 6pm

#

so despite feeling close i do have to put this away

serene lichen Jul 26, 2023, 6:08 PM

#

ok!

strange ridge Jul 26, 2023, 6:08 PM

#

did you short the pads on the back and use the DSI I2C for the config? (tahts the extended bus thing?)

serene lichen Jul 26, 2023, 6:09 PM

#

right yes, totally forgot to write that in the mail. Bridging the pads on the back and the using the i2c0 overlay for pins 44 and 45 to have i2c0 run on the DSI cable

#

and on the SAMD just empty loop so it doesn't interfere

#

and yea thats why I had to use extended bus as thats the easy way to make use of i2c0 with Blinka

strange ridge Jul 26, 2023, 6:11 PM

#

ok ok thank you!

#

i feel we got a lot workin here

#

ill maybe email tonite if i can try again 🙂

#

once i have a golden path i can usually work out how to automate it

#ah yea but those are inherent to the PI