KEXEC : No Display Out with "Warm Boot "

Robotics & Edge Computing Jetson Systems Jetson AGX Orin
,
1

Hello,

We are trying to run KEXEC.

But there is no display out on warm boot process.

Cold Boot => Display out on UEFI and Kernel both

Warm Boot => No display out with Kernel

Test Environment

  • AGX Orin Dev Kit
  • JetPack 6.2.1

Test process :

  1. Activate “CONFIG_PROC_KCORE” on defconfig
  2. Build the Kernel and Install L4T
  3. Run KEXEC

///////////////////////////////////////////////////////////////////////// save current device map
sudo cp /sys/firmware/fdt /tmp/current.dtb

///////////////////////////////////////////////////////////////////////// load the kernel image to RAM

sudo kexec -l /boot/Image \

       --initrd=/boot/initrd \\

       --dtb=/tmp/current.dtb \\

       --command-line="$(cat /proc/cmdline) KEXEC_TEST_SUCCESS=1"

cat /sys/kernel/kexec_loaded => check output value ‘1’

///////////////////////////////////////////////////////////////////////// run kexec

#!/bin/bash

sync

echo “<1> ========================================================” | sudo tee /dev/kmsg

echo “<1> [USER_MARKER] KEXEC 명령 시작합니다! (Start of Kexec)” | sudo tee /dev/kmsg

echo “<1> ========================================================” | sudo tee /dev/kmsg

# sudo kexec -e

sudo systemctl kexec

Please, check the logs.

001_kexec_with_kernel_log.txt (81.9 KB)

001_cold_reboot.txt (146.1 KB)

Is it possible to change the boot sequence so that ‘tegra-drm’ initializes properly instead of relying on ‘FB_SIMPLE’ during warm boot?

4

7.839943] nvgpu: 17000000.gpu nvgpu_nvhost_syncpt_init:122 [INFO] syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 10000
[ 7.839943]
[ 7.845490] nvgpu: 17000000.gpu gp10b_tegra_clks_control:190 [ERR] could not turn on clock 2
[ 7.846192] ------------[ cut here ]------------
[ 7.846194] gpc1clk already disabled
[ 7.846208] WARNING: CPU: 7 PID: 163 at drivers/clk/clk.c:969 clk_core_disable+0x23c/0x260
[ 7.846223] Modules linked in: nvgpu(O) governor_pod_scaling(O) host1x(O) mc_utils(O) nvmap(O) nvsciipc(O) drm fuse ip_tables x_tables ipv6 pwm_fan pwm_tegra tegra_bpmp_thermal tegra_xudc ucsi_ccg typec_ucsi typec nvme nvme_core phy_tegra194_p2u pcie_tegra194
[ 7.846243] CPU: 7 PID: 163 Comm: kworker/u24:3 Tainted: G O 5.15.148-tegra #1
[ 7.846246] Hardware name: NVIDIA NVIDIA Jetson AGX Orin Developer Kit/Jetson, BIOS 202601.0-28265983 02/12/2026
[ 7.846247] Workqueue: events_unbound async_run_entry_fn
[ 7.846254] pstate: 604000c9 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=–)
[ 7.846256] pc : clk_core_disable+0x23c/0x260
[ 7.846258] lr : clk_core_disable+0x23c/0x260
[ 7.846260] sp : ffff80000a4a38a0
[ 7.846261] x29: ffff80000a4a38a0 x28: ffff00008fd00000 x27: ffff00008fd00000
[ 7.846263] x26: ffff00008c0cfc80 x25: ffffc0bb23afd068 x24: ffffc0bb23b50540
[ 7.846265] x23: 0000000000000000 x22: ffffc0bb23b50300 x21: ffffc0bb23b50318
[ 7.846266] x20: ffff000083c88000 x19: ffff000083c88000 x18: ffffffffffffffff
[ 7.846268] x17: 5d5252455b202030 x16: ffffc0bb7126d120 x15: 6f635f736b6c635f
[ 7.846270] x14: 0000000000000000 x13: 0a64656c62617369 x12: 642079646165726c
[ 7.846272] x11: 656820747563205b x10: ffffc0bb72e996d8 x9 : ffff80000a4a38a0
[ 7.846273] x8 : 00000000ffffffff x7 : 656c626173696420 x6 : 79646165726c6120
[ 7.846275] x5 : ffff000fa0b2a9f0 x4 : 00000000fffff276 x3 : ffffc0bb72ee7958
[ 7.846278] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000832c2e80
[ 7.846280] Call trace:
[ 7.846282] clk_core_disable+0x23c/0x260
[ 7.846284] clk_disable+0x40/0x60
[ 7.846286] gp10b_tegra_clks_control+0x11c/0x150 [nvgpu]
[ 7.846433] ga10b_tegra_railgate+0x54/0x80 [nvgpu]
[ 7.846554] gk20a_pm_railgate+0x128/0x150 [nvgpu]
[ 7.846672] gk20a_pm_runtime_suspend+0xd8/0x100 [nvgpu]
[ 7.846785] pm_generic_runtime_suspend+0x44/0x70
[ 7.846791] genpd_runtime_suspend+0xb0/0x2d0
[ 7.846795] __rpm_callback+0x54/0x1b0
[ 7.846797] rpm_callback+0x8c/0xa0
[ 7.846799] rpm_suspend+0x114/0x650
[ 7.846802] __pm_runtime_suspend+0x68/0xe0
[ 7.846804] nvgpu_probe+0x7f0/0xa60 [nvgpu]
[ 7.846916] gk20a_probe+0x41c/0x6c0 [nvgpu]
[ 7.847027] platform_probe+0x74/0xf0
[ 7.847031] really_probe.part.0+0xac/0x320
[ 7.847032] __driver_probe_device+0xa4/0x170
[ 7.847034] driver_probe_device+0x58/0x1a0
[ 7.847035] __driver_attach_async_helper+0x54/0xc0
[ 7.847036] async_run_entry_fn+0x40/0x100
[ 7.847039] process_one_work+0x208/0x4d0
[ 7.847043] worker_thread+0x144/0x490
[ 7.847045] kthread+0x184/0x190
[ 7.847047] ret_from_fork+0x10/0x20
[ 7.847050] —[ end trace f08e3aecf342da8b ]—
[ 8.378053] systemd-journald[301]: Received client request to flush runtime journal.
[ 8.450037] ------------[ cut here ]------------
[ 8.450041] gpc1clk already unprepared

I am not quite sure about the meaning of your log.

The first one got above gpu crash which might lead to display fails to work.

But another one with “cold reboot” seems not reproduce any error. What is the exact scenario that would reproduce issue?