Issue with /dev/dri/ with no desktop running

oxoocoffee commented 5 months ago

Creating a bug report/issue

[x ] I have searched the existing open and closed issues

Required Information

DietPi version |

G_DIETPI_VERSION_CORE=9
G_DIETPI_VERSION_SUB=3
G_DIETPI_VERSION_RC=0
G_GITBRANCH='master'
G_GITOWNER='MichaIng'

Distro version | bookworm 0
Kernel version | Linux DietPi 6.6.20+rpt-rpi-2712 #1 SMP PREEMPT Debian 1:6.6.20-1+rpt1 (2024-03-07) aarch64 GNU/Linux
SBC model | RPi 5 Model B (aarch64) or (EG: RPi3)

Firmware/EEPROM |

BOOTLOADER: up to date
CURRENT: Sat Apr 20 10:53:30 UTC 2024 (1713610410)
LATEST: Sat Apr 20 10:53:30 UTC 2024 (1713610410)
RELEASE: default (/lib/firmware/raspberrypi/bootloader-2712/default)
        Use raspi-config to change the release.

Power supply used | 200W multi-port USB charger
SD card used | SanDisk 64G

Additional Information (if applicable)

Software title | Discovered DRI when researching ticket #7029
Was the software title installed freshly or updated/migrated? My test program which fais
Can this issue be replicated on a fresh installation of DietPi? Yes
Bug report ID | echo $G_HW_UUID

Steps to reproduce

Boot from a fresh image on rPi 5 and complete installation with our desktop selected.
Install build tools sudo apt install build-essential git cmake pkg-config libdrm-dev drm_info

ls /dev/dri/

drwxr-xr-x 2 root root        100 Apr 22 16:33 by-path
crw-rw---- 1 root video  226,   0 Apr 22 16:33 card0
crw-rw---- 1 root video  226,   1 Apr 22 16:33 card1
crw-rw---- 1 root render 226, 128 Apr 22 16:33 renderD128

Uncompress dri-test.tgz
cd dri-test/build && cmake .. && make

Expected behaviour

See a full screen with a test chess checker. I do not know if it fully work since DRI initialization fails.

Actual behaviour

Since my program terminates on a second call after open which is device.resources = drmModeGetResources(device.fd) with and error Failed to get DRM resources. Operation not supported

Extra details

groups produced tty dialout sudo audio video users input render netdev gpio i2c spi

MichaIng commented 5 months ago

KMS is enabled, right?

grep dtoverlay=vc4-kms-v3d /boot/config.txt

Did you start the test program from within a desktop session? I don't know what it is doing, where "device" and "fd" (file descriptor?) are coming from, but maybe it requires a context or input to know which of the 2 DRI devices it should use?

Did you test this with RPi OS Bookworm?

oxoocoffee commented 5 months ago

@MichaIng

Yes. All tests were performed with dtoverlay=vc4-kms-v3d

I did not stop from within a desktop. I have no desktop installed/running
I did not test rPI OS. I can try to download it and see. What specific version do you suggest?

oxoocoffee commented 4 months ago

While looking into this issue some more today I found this #5049

Change the config.txt file to dtoverlay=vc4-kms-v3d-pi4

MichaIng / DietPi