Nvmap dump

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin NX
1

我是jetpack5.1.2的版本，在使用中出现了这个dump，这可能是哪方面原因，该如何避免，是否是RT补丁的问题？
我们使用了nvjpeg用于编解码

[56609.469120] uvcvideo: Failed to set UVC probe control : -32 (exp. 48).
[56609.745957] uvcvideo: Failed to set UVC probe control : -32 (exp. 48).
[57311.980927] tegra-xusb 3610000.xhci: ERROR unknown event type 37
[58113.742278] zme_birdvision_: page allocation failure: order:7, mode:0x40dc0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO), nodemask=(null),cpuset=/,mems_allowed=0
[60380.712747] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000068
[60380.712758] Mem abort info:
[60380.712759]   ESR = 0x96000005
[60380.712760]   EC = 0x25: DABT (current EL), IL = 32 bits
[60380.712762]   SET = 0, FnV = 0
[60380.712763]   EA = 0, S1PTW = 0
[60380.712763] Data abort info:
[60380.712764]   ISV = 0, ISS = 0x00000005
[60380.712765]   CM = 0, WnR = 0
[60380.712766] user pgtable: 4k pages, 48-bit VAs, pgdp=000000025e314000
[60380.712768] [0000000000000068] pgd=00000003b3391003, p4d=00000003b3391003, pud=0000000000000000
[60380.712773] Internal error: Oops: 96000005 [#1] PREEMPT_RT SMP
[60380.712777] Modules linked in: nf_conntrack_netlink nfnetlink br_netfilter lzo_rle lzo_compress zram overlay ramoops reed_solomon loop bnep hid_sensor_gyro_3d hid_sensor_accel_3d hid_sensor_trigger industrialio_triggered_buffer kfifo_buf hid_sensor_custom hid_sensor_iio_common snd_soc_tegra210_ope snd_soc_tegra186_dspk snd_soc_tegra186_asrc snd_soc_tegra186_arad snd_soc_tegra210_iqc snd_soc_tegra210_mvc snd_soc_tegra210_afc snd_soc_tegra210_dmic snd_soc_tegra210_adx ch341 usbserial hid_sensor_hub snd_soc_tegra210_amx rtk_btusb btusb btrtl btbcm snd_soc_tegra210_i2s btintel snd_soc_tegra210_mixer snd_soc_tegra210_admaif snd_soc_tegra210_sfc snd_soc_tegra_pcm aes_ce_blk crypto_simd cryptd aes_ce_cipher ghash_ce sha2_ce sha256_arm64 sha1_ce snd_soc_spdif_tx snd_soc_tegra_machine_driver ip6table_filter fusb301 ip6_tables xt_state snd_soc_tegra210_adsp xt_conntrack iptable_filter snd_soc_tegra_utils tegra_bpmp_thermal snd_soc_simple_card_utils nv_imx219 xt_MASQUERADE userspace_alert nvadsp
[60380.712838]  xt_addrtype snd_hda_codec_hdmi iptable_nat snd_soc_tegra210_ahub tegra210_adma nf_nat snd_hda_tegra snd_hda_codec nf_conntrack rtl8822ce snd_hda_core nf_defrag_ipv6 nf_defrag_ipv4 cfg80211 libcrc32c r8168 spi_tegra114 nvidia(OE) binfmt_misc uvcvideo videobuf2_vmalloc ina3221 pwm_fan nvgpu nvmap ip_tables x_tables [last unloaded: mtd]
[60380.712863] CPU: 0 PID: 147821 Comm: right_loop Tainted: G           OE     5.10.120-rt70-tegra #11
[60380.712866] Hardware name: Unknown NVIDIA Orin NX Developer Kit start dev0220/NVIDIA Orin NX Developer Kit start dev0220, BIOS 4.1-33958178 08/01/2023
[60380.712869] pstate: 20400009 (nzCv daif +PAN -UAO -TCO BTYPE=--)
[60380.712871] pc : nvmap_free_handle_from_fd+0xd0/0x260 [nvmap]
[60380.712891] lr : nvmap_free_handle_from_fd+0x80/0x260 [nvmap]
[60380.712900] sp : ffff80002717bd60
[60380.712900] x29: ffff80002717bd60 x28: ffff081a8026bb00
[60380.712902] x27: 0000000000000000 x26: 0000000000000000
[60380.712904] x25: 0000000000000000 x24: 0000000000000000
[60380.712906] x23: 0000000000000000 x22: 0000000000000000
[60380.712908] x21: 00000000800000e1 x20: ffff08190df9c100
[60380.712910] x19: ffff0819b59acc00 x18: 0000000000000000
[60380.712911] x17: 0000000000000000 x16: ffffaf6e9ea8f140
[60380.712913] x15: 0000000000000000 x14: 0000000000000000
[60380.712915] x13: 0000000000000000 x12: 0000000000000000
[60380.712916] x11: ffff80002717bce8 x10: 0000000000000002
[60380.712918] x9 : 0000000000000001 x8 : 0000000000000238
[60380.712920] x7 : 0000000346cd2000 x6 : 0000000000000018
[60380.712921] x5 : ffffaf6e5b03b8a0 x4 : 0000000000000003
[60380.712923] x3 : 0000000000000002 x2 : ffff0819b59ace00
[60380.712925] x1 : 0000000000000003 x0 : 0000000000000000
[60380.712927] Call trace:
[60380.712928]  nvmap_free_handle_from_fd+0xd0/0x260 [nvmap]
[60380.712937]  nvmap_ioctl_free+0x44/0x90 [nvmap]
[60380.712946]  nvmap_ioctl+0x1c4/0x500 [nvmap]
[60380.712955]  __arm64_sys_ioctl+0xac/0xf0
[60380.712963]  el0_svc_common.constprop.0+0x80/0x1d0
[60380.712974]  do_el0_svc+0x38/0xb0
[60380.712978]  el0_svc+0x1c/0x30
[60380.712985]  el0_sync_handler+0xa8/0xb0
[60380.712988]  el0_sync+0x16c/0x180
[60380.712995] Code: b40000c2 f9400440 f00000a5 912280a5 (f9403404)
[60380.712999] ---[ end trace 0000000000000002 ]---
[60381.148792] usb 2-1.3.1: Process 148118 (cameras_streams) called USBDEVFS_CLEAR_HALT for active endpoint 0x82
[60381.149147] usb 2-1.2: Process 148234 (cameras_streams) called USBDEVFS_CLEAR_HALT for active endpoint 0x84
3

Hi 1031150349,

Are you using the devkit or custom board for Orin NX?

Please share the detailed steps how you reproduce this issue.

4

I’m using a board I made myself, and it seems that using nvjpeg is causing the error, which we are currently investigating.

5

Is the issue also happening on the devkit board?
Or it is specific to your custom board?

Please share the steps to use nvjpeg and how to cause the error.

6

This is what C++ engineers say：
orin NX uses multithreading to capture 128072030fps streams from four USB cameras in mjpeg format. The single-frame image format jpeg is captured and the decodeToFd() method in the jetson_multimedia_api module sample 06_jpeg_decode is used for nvjepg decoding. After the program runs for a while, the above problem will occur.

we use custom board.

7

Do you have the devkit to verify if it would hit the same issue?

Please also share the detailed steps for us to verify it locally.

8

I’ve been busy with other things recently, and I plan to test this process tomorrow to see if there will be any problems with the devkit.

10

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