rockowitz
ddcutil detect tries to open /sys/bus/i2c/devices/i2c--1
Hi, when I've tried to use ddcutil detect I've got the following output:
Unable to open directory /sys/bus/i2c/devices/i2c--1: No such file or directory
Device /dev/i2c-255 does not exist. Error = ENOENT(2): No such file or directory
/sys/bus/i2c buses without /dev/i2c-N devices: /sys/bus/i2c/devices/i2c-255
Driver i2c_dev must be loaded or builtin
See https://www.ddcutil.com/kernel_module
Invalid display
I2C bus: /dev/i2c-5
DRM connector: card0-eDP-1
EDID synopsis:
Mfg id: CMN - Chimei Innolux Corporation
Model:
Product code: 4977 (0x1371)
Serial number:
Binary serial number: 0 (0x00000000)
Manufacture year: 2016, Week: 11
DDC communication failed
This is an eDP laptop display. Laptop displays do not support DDC/CI.
Phantom display
I2C bus: /dev/i2c-6
DRM connector: card0-DP-1
EDID synopsis:
Mfg id: PHL - Philips Consumer Electronics Company
Model: PHL 275E2F
Product code: 49722 (0xc23a)
Serial number: UHB2141009681
Binary serial number: 9681 (0x000025d1)
Manufacture year: 2021, Week: 41
DDC communication failed
Use non-phantom device bus /dev/i2c-7
Display 1
I2C bus: /dev/i2c-7
DRM connector: card0-DP-2
EDID synopsis:
Mfg id: PHL - Philips Consumer Electronics Company
Model: PHL 275E2F
Product code: 49722 (0xc23a)
Serial number: UHB2141009681
Binary serial number: 9681 (0x000025d1)
Manufacture year: 2021, Week: 41
VCP version: 2.2
The output of ls /sys/bus/i2c/devices is :
1-0036 1-0037 1-0050 i2c-0 i2c-1 i2c-2 i2c-3 i2c-4 i2c-5 i2c-6 i2c-7 i2c-MSFT0001:00
I guess that i2c-MSFT0001:00 device is causing that issue in the function https://github.com/rockowitz/ddcutil/blob/b54f62a/src/i2c/i2c_sysfs.c#L1244 ( where i2c_name_to_busno returns -1 ).
I'm using the Gnome Brightness control using ddcutil extension which stopped working recently. In the logs I get a similar message:
ddcutil reading display status for bus: 4 is: Unable to open directory /sys/bus/i2c/devices/i2c--1: No such file or directory
Device /dev/i2c-255 does not exist. Error = ENOENT(2): No such file or directory
/sys/bus/i2c buses without /dev/i2c-N devices: /sys/bus/i2c/devices/i2c-255
Driver i2c_dev must be loaded or builtin
See https://www.ddcutil.com/kernel_module
VCP D6 ERR
The i2c_dev driver is loaded:
$ lsmod | grep i2c_dev
i2c_dev 24576 0
Running the command to change the brightness does work:
$ ddcutil setvcp 10 15
@firmero Thank you for the precise bug report,. Branch 1.4.0-dev contains a set of commits that address the problem. Let me know if this works for you.
Out of curiosity, what is the i2c-MSFT0001:00 device?
It appears that you have a single external monitor that is detected at 2 different i2c addresses (a phantom display). How is the display connected? What is the laptop model? Please execute ddcutil environment --verbose and submit the output as an attachment of some sort.
@julienfalque I was unaware of the Gnome Brightness Control. If you can, please install branch 1.4.0-dev and let me know if it addresses the problem.
@rockowitz thanks, changes resolved my issue. Ddcutil detect's output looks good to me: ddcutil-detect.txt.
Device i2c-MSFT0001:00 is probably my touchpad (driver=i2c_hid_acpi). My setup contains a notebook display + external monitor connected with my ntb via the docking station (Dell Dock WD15). My environment: ddcutil-environment.txt
Btw, Gnome Brightness Control extension stopped working, so I got here. That extension relies on one line output of getvcp --brief D6 --bus $BUS (then basic split on " " is used), but it was messed out with Unable to open directory error.
@firmero Thank you for the environment output. It helps me get a better understanding on how /sys is used for MST, so that I can more definitively discard the "phantom" display. For a discussion of the problem, see issue Same MST display appears as 2 /dev/i2c devices on the drm/intel list.
I just tried 1.4.0-dev (d966330086caca93fdd71a8be43c77a903d1785f). It seems better but I think it's still not working as expected:
$ ddcutil detect
Invalid display
I2C bus: /dev/i2c-3
DRM connector: card0-eDP-1
EDID synopsis:
Mfg id: SHP - Sharp Corporation
Model:
Product code: 5182 (0x143e)
Serial number:
Binary serial number: 0 (0x00000000)
Manufacture year: 2015, Week: 40
DDC communication failed
This is an eDP laptop display. Laptop displays do not support DDC/CI.
Display 1
I2C bus: /dev/i2c-4
DRM connector: card0-DP-1
EDID synopsis:
Mfg id: GBT - UNK
Model: AORUS FI32U
Product code: 12802 (0x3202)
Serial number: 21400B000112
Binary serial number: 16843009 (0x01010101)
Manufacture year: 2021, Week: 40
VCP version: Detection failed
Changing the brightness still works but with error output and exit status:
$ ddcutil setvcp 10 15
VCP (aka MCCS) version for display is undetected or less than 2.0. Output may not be accurate.
Setting value failed for feature 10, rc=DDCRC_ALL_RESPONSES_NULL(-3011): all tries returned DDC Null Message
$ echo $?
1
Also, Gnome extension still does not work.
Actually the behavior for ddcutil setvcp 10 15 is the same on 1.3.0 with more errors, I forgot to report that in my first message:
$ ddcutil setvcp 10 15
Unable to open directory /sys/bus/i2c/devices/i2c--1: No such file or directory
Device /dev/i2c-255 does not exist. Error = ENOENT(2): No such file or directory
/sys/bus/i2c buses without /dev/i2c-N devices: /sys/bus/i2c/devices/i2c-255
Driver i2c_dev must be loaded or builtin
See https://www.ddcutil.com/kernel_module
VCP (aka MCCS) version for display is undetected or less than 2.0. Output may not be accurate.
Setting value failed for feature 10, rc=DDCRC_ALL_RESPONSES_NULL(-3011): all tries returned DDC Null Message
@julienfalque Please run ddcutil interrogate (as root if necessary) and submit the output as an attachment. Thank you.
The error your seeing is from reading the value after the write to verity that it has been set correctly. You can disable verification using option --noverify.
More generally, there's something a bit marginal about your monitor, which can be seen in the retry data in the interrogate output. Try using option --sleep-multiplier 2.0. Also if ddcutil commands are being issued in rapid succession from a script, putting a short sleep, say .5 seconds, between each ddcutil call can help.
As to the Gnome Brightness Extension, it's a "screen scraping" application that's not handling unexpected ddcutil output. When I return from vacation I'll look into reducing the error messages in the --brief output that the Brightness extension uses, but a screen scraping application should extension should be able to handle unexpected program output in some way.
Finally, as noted previously, the error with double hyphens in /dev/i2c names, i.e. /dev/i2c--1", which exists in 1.3.0, has been fixed in 1.4.0-dev. If it still occurs in in 1.4.0-dev that should be reported.
I confirm increasing sleep multiplier fixes the remaining issue. Everything looks good to me, thank you!
Worth noting that all this Unable to open directory output (all 5 lines of it) goes to stdout no matter that --terse or --syslog is specified. Hope this is addressed in 1.4.0-dev too.
I am getting this too. After the latest Arch update to Version 1.4.0.r0.g670b6f36-1 ddcutil-git stopped working altogether (displays and features detected but all write attempts did not change anything and then, of course, "Verification failed"). So I downgraded ddcutil to 1.3.0 stable. Now value write changes do work again, but I get a lot of gibberish on every invocation:
Unable to open directory /sys/bus/i2c/devices/i2c--1: No such file or directory
Unable to open directory /sys/bus/i2c/devices/i2c--1: No such file or directory
Unable to open directory /sys/bus/i2c/devices/i2c--1: No such file or directory
Device /dev/i2c-255 does not exist. Error = ENOENT(2): No such file or directory
/sys/bus/i2c buses without /dev/i2c-N devices: /sys/bus/i2c/devices/i2c-255
Driver i2c_dev must be loaded or builtin
And yes, the driver module is loaded, of course :-)
All the output you are seeing are the result of the mis-parsing of /dev/i2c devices with 2 hyphens in the name, which was fixed in release 1.4.0.
As a first step to diagnosing the problem, please try executing a setvcp command with option --trace all. Also, please run ddcutil interrogate so that I know your configuration. And submit the output as attachments or some other method that is not in the text of your reply. Thank you.