Display Output Issue on i.MX95 – LVDS Connection Not Working

[DusanStanisic99]

Hello everyone,

We have successfully built and flashed a Yocto image for the NXP i.MX95LPD5EVK-19CM board. According to the Quick Start Guide, the display should be connected via the J14 MIPI-DSI connector using the IMX-MIPI-HDMI adapter. However, we currently only have access to the IMX-LVDS-HDMI adapter, which we previously used with the NXP i.MX8 platform.

We noticed that the i.MX95 board includes two LVDS connectors. We attempted to connect our display using both, but the board does not appear to recognize the connected monitor. (We’ve attached photos for reference.)

Our main question is:
Does the i.MX95 board supports LVDS display output, or is this functionality possibly disabled in the current device tree configuration? Alternatively, is MIPI the only supported display interface for this board, requiring us to obtain the IMX-MIPI-HDMI adapter?

Any guidance or clarification would be greatly appreciated.

Best regards,
Dušan

[DusanStanisic99]

Here are some additional logs from dmesg:

root@imx95evk:~# dmesg -w | grep drm
[    2.977422] [drm:drm_bridge_attach] *ERROR* failed to attach bridge /soc/dsi@4acf0000 to encoder None-39: -19
[    2.987370] [drm:drm_bridge_attach] *ERROR* failed to attach bridge /soc/syscon@4b010000/bridge@8/ports/port@0 to encoder None-39: -19
[    3.006505] [drm:drm_bridge_attach] *ERROR* failed to attach bridge /soc/bridge@4b0d0000/channel@0 to encoder None-39: -19
[    3.006520] imx95-dpu 4b400000.display-controller: [drm] *ERROR* failed to attach bridge to encoder for stream0: -19
[    6.895845] systemd[1]: Starting Load Kernel Module drm...
^C
root@imx95evk:~# dmesg -w | grep gpu
[    2.412169] mali 4d900000.gpu: Kernel DDK version r50p0-00eac0
[    2.419577] mali 4d900000.gpu: Register LUT 000a0a00 initialized for GPU arch 0x000a0c07
[    2.433046] mali 4d900000.gpu: r0p0 status 1 not found in HW issues table;
[    2.445219] mali 4d900000.gpu: falling back to closest match: r0p0 status 0
[    2.461651] mali 4d900000.gpu: Execution proceeding normally with fallback match
[    2.469057] mali 4d900000.gpu: GPU identified as 0x4 arch 10.12.7 r0p0 status 0
[    2.481551] mali 4d900000.gpu: CSF_GPU_RESET_TIMEOUT is capped from 12400ms to 4500ms
[    2.489391] mali 4d900000.gpu: KBASE_PRFCNT_ACTIVE_TIMEOUT is capped from 24000ms to 4500ms
[    2.489396] mali 4d900000.gpu: KBASE_AS_INACTIVE_TIMEOUT is capped from 24000ms to 4500ms
[    2.511041] mali 4d900000.gpu: CSF_FIRMWARE_STOP_TIMEOUT is capped from 24000ms to 4500ms
[    2.519312] mali 4d900000.gpu: No priority control manager is configured
[    2.526015] mali 4d900000.gpu: Large page support was disabled at compile-time!
[    2.526042] mali 4d900000.gpu: No memory group manager is configured
[    2.526060] mali 4d900000.gpu: Protected memory allocator not available
[    2.557911] mali 4d900000.gpu: Using configured power model mali-tvax-power-model, and fallback mali-simple-power-model
[    2.575151] mali 4d900000.gpu: OPP 0 : opp_freq=1000000000 core_mask=1
[    2.589295] mali 4d900000.gpu: OPP 1 : opp_freq=800000000 core_mask=1
[    2.589301] mali 4d900000.gpu: OPP 2 : opp_freq=500000000 core_mask=1
[    2.596639] mali 4d900000.gpu: * MALI kbase_mmap_min_addr compiled to CONFIG_DEFAULT_MMAP_MIN_ADDR, no runtime update possible! *
[    2.635192] mali 4d900000.gpu: Probed as mali0
[    3.772845] debugfs: Directory 'gpu' with parent 'pm_genpd' already present!

[Bio_TICFSL]

Hello,

Some modification in the dtb are needed to have LVDS working on MX95, however the latest linux BSP included it on it , please check it.

Your fails is more like missing GPU,  please use the full-image in yocto and see whats happens.

Regards

[DusanStanisic99]

Hi, 

Since we are using an HMI-LVDS-HDMI adapter, we managed to redirect output to our standard HDMI display's full HD resolution by using a combination of dtb (imx95-19x19-evk-it6263-lvds-dual.dtb) and overlay (imx95-19x19-evk-it6263-lvds-dual.dtbo) instead of the default one.

However, when we plugged in our use-case-based HDMI display resolution of 1920x720, we had a failure in our driver (picture will be added), and also a failure in the Weston service (log will be provided).

This suggests to me that maybe some custom resolutions are not supported by default. Could you provide us with directions on what we should look into, or maybe where the changes should be made?

Weston service logs:
root@imx95-19x19-lpddr5-evk:~# cat /run/user/0/weston.log
Date: 2025-05-21 UTC
[12:03:24.176] weston 12.0.4
https://wayland.freedesktop.org
Bug reports to: https://gitlab.freedesktop.org/wayland/weston/issues/
Build: 10.0.0-1228-gc3c1f9ff+
[12:03:24.177] Command line: /usr/bin/weston --log=/run/user/0/weston.log --modules=systemd-notify.so
[12:03:24.177] OS: Linux, 6.6.36-lts-next-g27c987054d68, #1 SMP PREEMPT Wed Sep 4 08:22:45 UTC 2024, aarch64
[12:03:24.177] Flight recorder: enabled
[12:03:24.177] Using config file '/etc/xdg/weston/weston.ini'
[12:03:24.177] Output repaint window is 16 ms maximum.
[12:03:24.177] Loading module '/usr/lib/libweston-12/drm-backend.so'
[12:03:24.193] initializing drm backend
[12:03:24.193] Trying libseat launcher...
[12:03:24.193] [c/debug/seatd/0.8.0/libseat/backend/seatd.c:64] Could not connect to socket /run/seatd.sock: No such file or directory
[12:03:24.193] [c/debug/seatd/0.8.0/libseat/libseat.c:76] Backend 'seatd' failed to open seat, skipping
[12:03:24.199] [c/debug/seatd/0.8.0/libseat/libseat.c:73] Seat opened with backend 'logind'
[12:03:24.199] libseat: session control granted
[12:03:24.205] using /dev/dri/card0
[12:03:24.205] DRM: supports atomic modesetting
[12:03:24.205] DRM: supports GBM modifiers
[12:03:24.205] DRM: does not support async page flipping
[12:03:24.205] DRM: supports picture aspect ratio
[12:03:24.205] Loading module '/usr/lib/libweston-12/g2d-renderer.so'
[12:03:24.236] event0 - scmi_dev.11: is tagged by udev as: Keyboard
[12:03:24.236] event0 - scmi_dev.11: device is a keyboard
[12:03:24.243] event6 - WM8962 Beep Generator: not tagged as supported input device
[12:03:24.244] event6 - not using input device '/dev/input/event6'
[12:03:24.254] event1 - Logitech USB Keyboard: is tagged by udev as: Keyboard
[12:03:24.254] event1 - Logitech USB Keyboard: device is a keyboard
[12:03:24.264] event2 - Logitech USB Keyboard Consumer Control: is tagged by udev as: Keyboard
[12:03:24.265] event2 - Logitech USB Keyboard Consumer Control: device is a keyboard
[12:03:24.274] event3 - Logitech USB Keyboard System Control: is tagged by udev as: Keyboard
[12:03:24.274] event3 - Logitech USB Keyboard System Control: device is a keyboard
[12:03:24.284] event4 - Logitech USB Optical Mouse: is tagged by udev as: Mouse
[12:03:24.285] event4 - Logitech USB Optical Mouse: device is a pointer
[12:03:24.359] event5 - HID 0712:000a: is tagged by udev as: Touchscreen
[12:03:24.360] event5 - HID 0712:000a: device is a touch device
[12:03:24.379] libinput: configuring device "scmi_dev.11".
[12:03:24.379] libinput: configuring device "Logitech USB Keyboard".
[12:03:24.379] libinput: configuring device "Logitech USB Keyboard Consumer Control".
[12:03:24.379] libinput: configuring device "Logitech USB Keyboard System Control".
[12:03:24.379] libinput: configuring device "Logitech USB Optical Mouse".
[12:03:24.379] Touchscreen - HID 0712:000a - /sys/devices/platform/soc/4c200000.usb/ci_hdrc.0/usb1/1-1/1-1.3/1-1.3:1.0/0003:0712:000A.0005/input/input11/event5
[12:03:24.379] libinput: configuring device "HID 0712:000a".
[12:03:24.379] input device event5 has no enabled output associated (none named), skipping calibration for now.
[12:03:24.469] DRM: head 'HDMI-A-1' found, connector 40 is connected, EDID make 'KTC', model 'WaveShare', serial '1'
Supported EOTF modes: SDR
[12:03:24.470] Registered plugin API 'weston_drm_output_api_v1' of size 40
[12:03:24.470] Color manager: no-op
[12:03:24.470] Loading module '/usr/lib/libgbm.so'
[12:03:24.470] Failed to load module: /usr/lib/libgbm.so: cannot open shared object file: No such file or directory
[12:03:24.471] no available modes for HDMI-A-1 (this is related to the picture above)
[12:03:24.471] Cannot configure an output using weston_drm_output_api.
[12:03:24.472] event0 - scmi_dev.11: device removed
[12:03:24.474] event1 - Logitech USB Keyboard: device removed
[12:03:24.476] event2 - Logitech USB Keyboard Consumer Control: device removed
[12:03:24.478] event3 - Logitech USB Keyboard System Control: device removed
[12:03:24.480] event4 - Logitech USB Optical Mouse: device removed
[12:03:24.481] event5 - HID 0712:000a: device removed

Best regards,

Dušan

[Kabil]

Hi @DusanStanisic99,
I am also facing the same issue with the i.MX95 LVDS display output. Could you please let me know if you were able to resolve this problem, and if so, how you fixed it?

Thanks in advance.

[DusanStanisic99 · Accepted solution]

Hi @Kabil,

We managed to resolve the issue, but we haven’t updated the post online yet. The solution was to change the default DTB that gets loaded to: imx95-19x19-evk-it6263-lvds1.dtb

You can use serial communication to stop execution at U-Boot, and then run:

• setenv fdtfile imx95-19x19-evk-it6263-lvds1.dtb
• saveenv
• boot

With this change, the display output will be permanently set to LVDS.

Hope this helps!

[Kabil]

Hi @DusanStanisic99,

I tried the steps you mentioned and updated the DTB to imx95-19x19-evk-it6263-lvds1.dtb. However, I’m now facing the following issue during boot:

root@imx95a1evk:~# dmesg | grep it6263
[ 3.514817] it6263 2-004c: probe with driver it6263 failed with error -5

root@imx95a1evk:~# dmesg | grep drm
[ 3.611604] imx95-ldb 4b0c0000.syscon:ldb@4: [drm:ldb_init_helper] *ERROR* invalid channel node address: 1
[ 6.837909] systemd[1]: Starting Load Kernel Module drm...

root@imx95a1evk:~# ls /sys/class/drm/
version

Do you have any suggestions on how to resolve this probe failure and DRM channel node error?
Thanks in advance for your help!