Jetson Nano Devkit\Custom Carry board HDMI 2.0(2160P/60Hz) SI test problem

Autonomous Machines Jetson & Embedded Systems Jetson Nano
1

Hi NV Partner,

We are testing Nano HDMI 2.0 signals,and found both the Devkit board and our designed carrier board seem run INCORRECT frequency as 133MHz when HDMI output set 2160P/60Hz, at the same time the kernel message show display pix-clock is 533MHz, this phenomenon of out of sync doesn’t make sense.

For this reason , we could not go on HDMI testing, any suggesting about this problem?

2

The same problem with me, is there anyone help me?

4

Hi, did you follow the test application note in DLC first? There are some tuning request in it. https://developer.nvidia.com/jetson-nano-interface-tuning-and-compliance-guide-application-note

5

Yes,we followed the NV test application note;for now ,we have tested HDMI 1.4 4K/30Hz. Our oscilloscope is N5399C which is applicable for HDMI 2.0 test.
Any suggestions about the Topic question?

7

Did you follow the guide to change the resolution?Welcome — Jetson Linux<br/>Developer Guide 34.1 documentation

The frequency should be set automatically if supported resolution is selected. Is 133MHz the value you measured on scope?

8

We have tested 2160P/1080P/720P/480P (HDMI1.4) successfully, only the HDMI2.0 2160P/60Hz failed to test.
Yes, the 133MHz is measured on scope while the kernel message show 533MHz.

12

1, Yeah, we know Nano hardware capability can display real 4K@60, but now it seems Nano hardware not capable for 4K@60 for what we have measured HDMI 2.0 SI by scope,we want to know how Nano 4K@60 worked. By the way, our customized carrier board have only one display port–HDMI.
2,Yes, 4K@60 depends on the monitor.
You can refer below picture for pclk 533MHz, is 533MHz abnormal?

微信图片_20200511165238
微信图片_202005111652381440×1080 180 KB

14

,Yes, 4K@60 depends on the monitor.
You can refer below picture for pclk 533MHz, is 533MHz abnormal?

Could you paste the full log instead of a picture?

Also, need to check below node

sudo -s
sudo cat /sys/kernel/debug/tegradc.0/mode

#Add below option in /etc/X11/xorg.conf, ‘Section “Device”’. Verbose mode logging will be shown in /var/log/Xorg.0.log
Section “Device”
…
Option “ModeDebug”
…
EndSection

15

FYI.

cat mode
pclk: 533250000
h_ref_to_sync: 1
v_ref_to_sync: 2
h_sync_width: 32
v_sync_width: 5
h_back_porch: 80
v_back_porch: 54
h_active: 3840
v_active: 2160
h_front_porch: 48
v_front_porch: 3
flags: 0x1
stereo_mode: 0
avi_m: 0x2
vmode: 0x200000

16

Please also share the full dmesg and the log in xorg.0.log.

17

Also, we need to check this node.

sudo -s
cat /sys/kernel/debug/clk/pll_d2/clk_rate

18

What release are you using?

19

FYI.
dmesg.log (92.5 KB) Xorg.0.log (11.1 KB)

20

cat /sys/kernel/debug/clk/pll_d2/clk_rate
533250000

21

The release we are using is 32.2.1

22

You didn’t enable the verbose log in xorg.0.log. Please follow the instruction in my previous comments.

23

hi Wayne

FYI-Xorg.0.log (203.4 KB)

25

Could you also dump cat /sys/kernel/debug/clk/pll_d2/pto_rate?

26

root@test-desktop:/sys/kernel/debug/clk/pll_d2# cat clk_rate
533250000
root@test-desktop:/sys/kernel/debug/clk/pll_d2# cat pto_rate
533238000

27

Hi Wayne,

All logs you need have sent to you.
As you said the software may not definitely support HDMI 2.0, in another word, the Nano couldn’t support EXACT HDMI 2.0 3840x2160@60Hz, is it right?
So, we could not go on our HDMI SI test, Could you please give some other suggestions about 3840x2160@60Hz test?