Fail to shutdown with PHILIPS 241B8Q

I’m working on Orin NX 8GB + Xavier dev kit in R35.4.1.

I connect FHILIPS 241B8Q (display), Low speed USB storage and Uart debug cable.
The following error occurred when shutting down while writing to storage.
And reboot by watchdog.

[  115.312258] CPU: 3 PID: 1458 Comm: Xorg Tainted: G           O      5.10.120-tegra #1
[  115.320313] Hardware name: Unknown NVIDIA Orin NX Developer Kit/NVIDIA Orin NX Developer Kit, BIOS 4.1-33958178 08/01/2023
[  115.331678] Call trace:
[  115.334204]  dump_backtrace+0x0/0x1d0
[  115.337962]  show_stack+0x30/0x40
[  115.341373]  dump_stack+0xd8/0x138
[  115.344944]  os_dump_stack+0x18/0x20 [nvidia]
[  115.349477]  tlsEntryGet+0x130/0x138 [nvidia]
[  115.354008]  gpumgrGetSomeGpu+0x7c/0x90 [nvidia]
[  115.358795]  threadPriorityStateFree+0x234/0x2a0 [nvidia]
...

err.log (318.7 KB)

I suspect that the following are the causes.

  1. GDM is Stopped by shutdown process.
  2. Writing to low speed storage causes a wait for shutdown.
  3. FHILIPS 241B8Q enters suspend mode. And HPD goes from LOW to HIGH.
  4. The error happens due to an HPD interrupt while the GDM is stopped.
    WARNING occurs when plugin/out HDMI in OrinNX - #42 by shinichiro.adachi
  5. reboot by watchdog.

This occurred on both OrinNX 8GB and OrinNano 4GB.

I think this problem is caused by the HPD changing when GUI is stopped.
When will the bug about HPD be fixed?

Looks duplicated with Orin NX HDMI kernel panic when shutdown and connect Dell monitor - #49 by jack_lan

1 Like

Thanks.
They are very similar, just different monitors.

I try to modify 0x2212000 before shutting down.
But, it fails.

root@contec-desktop:~# devmem2 0x2212000 h 0xD
/dev/mem opened.
Memory mapped at address 0xffff92f91000.
[  108.401208] CPU:0, Error: cbb-fabric@0x13a00000, irq=25
[  108.406601] **************************************
[  108.411529] CPU:0, Error:cbb-fabric, Errmon:2
[  108.416010]    Error Code            : SLAVE_ERR
[  108.419872]
[  108.421400]    Error Code            : SLAVE_ERR
[  108.425253]    MASTER_ID             : CCPLEX
[  108.428747]    Address               : 0x2212000
[  108.432337]    Cache                 : 0x0 -- Device Non-Bufferable
[  108.437526]    Protection            : 0x2 -- Unprivileged, Non-Secure, Data Access
[  108.444493]    Access_Type           : Read
[  108.447980]    Access_ID             : 0x11
[  108.447982]    Fabric                : cbb-fabric
[  108.454874]    Slave_Id              : 0x3b
[  108.458093]    Burst_length          : 0x0
[  108.461580]    Burst_type            : 0x1
[  108.464894]    Beat_size             : 0x1
[  108.468113]    VQC                   : 0x0
[  108.470889]    GRPSEC                : 0x7e
[  108.473938]    FALCONSEC             : 0x0
[  108.477157]  **************************************
[  108.482222] WARNING: CPU: 0 PID: 2775 at drivers/soc/tegra/cbb/tegra234-cbb.c:577 tegra234_cbb_isr+0x130/0x170
[  108.492751] ---[ end trace 495bf1102cae1e03 ]---
Value at address 0x2212000 (0xffff92f91000): 0xFFFF
[  108.497931] CPU:0, Error: cbb-fabric@0x13a00000, irq=25
[  108.503300] **************************************
[  108.508221] CPU:0, Error:cbb-fabric, Errmon:2
[  108.512695]    Error Code            : SLAVE_ERR
[  108.516543]    Overflow              : Multiple SLAVE_ERR
[  108.521022]
[  108.522549]    Error Code            : SLAVE_ERR
[  108.526397]    MASTER_ID             : CCPLEX
[  108.529889]    Address               : 0x2212000
[  108.533470]    Cache                 : 0x0 -- Device Non-Bufferable
[  108.538656]    Protection            : 0x2 -- Unprivileged, Non-Secure, Data Access
[  108.545615]    Access_Type           : Write
[  108.549199]    Access_ID             : 0x0
[  108.549201]    Fabric                : cbb-fabric
[  108.556005]    Slave_Id              : 0x3b
[  108.559223]    Burst_length          : 0x0
[  108.562710]    Burst_type            : 0x1
[  108.566026]    Beat_size             : 0x1
[  108.569244]    VQC                   : 0x0
[  108.572022]    GRPSEC                : 0x7e
[  108.575072]    FALCONSEC             : 0x0
[  108.578296]  **************************************
[  108.583343] WARNING: CPU: 0 PID: 2775 at drivers/soc/tegra/cbb/tegra234-cbb.c:577 tegra234_cbb_isr+0x130/0x170
[  108.593837] ---[ end trace 495bf1102cae1e04 ]---
Written 0xD; readback 0xFFFF

What is the recommended way to change register values?

Oh, that is because the firewall setting (the SCR file you modified before flashing your board) blocks the CCPLEX to access the register.

Then there is no other method. You could try to use the default SCR setting as rel-35.3.1 and it shall be able to access again.

My fix is old.

<bootloader/t186ref/BCT/tegra234-mb2-bct-scr-p3767-0000.dts >
[old-fix]
        reg@322 { /* GPIO_M_SCR_00_0 */
            exclusion-info = <2>;
            value = <0x38008080>;
        };
[new-fix]
        reg@322 { /* GPIO_M_SCR_00_0 */
            exclusion-info = <2>;
            value = <0x38009696>;
        };

I reflash. it works.
Thanks.

1 Like

Hi, @WayneWWW
By the way.
When will the bug about HPD be fixed?
Next release?

Target to be jetpack6.

1 Like

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.