Hailo8 dmesg errors with new kernel

General Discussion
,
1

Hello,

recently we updated kernel from

Ubuntu 24.04.2 LTS, Linux test 6.8.0-41-generic #41-Ubuntu SMP PREEMPT_DYNAMIC Fri Aug 2 20:41:06 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux

to

Ubuntu 24.04.3 LTS, Linux TA120-C001 6.14.0-29-generic #29~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Aug 14 16:52:50 UTC 2 x86_64 x86_64 x86_64 GNU/Linux

we started to get warnings in dmesg.

I tried to build drivers from source, but that did not help.

We are using drivers version 4.20.0. but same thing happend with 4.22.0 as well.

We are going to need newer kernel for CPU driver compatibility.

I could not attach full dmesg log, so here is only the last bit, full log had ~2,5k lines, I think the error produced by our test was still the same only repeated many times anyway.

How do we fix this please? Have anyone run into this problem?

Thank you for any suggestion,

Daniel

[  514.793444] ------------[ cut here ]------------
[  514.793446] WARNING: CPU: 0 PID: 2245 at include/linux/rwsem.h:80 find_vma+0x5b/0x70
[  514.793453] Modules linked in: tcp_diag raw_diag inet_diag cfg80211 sch_fq_codel xe drm_gpuvm gpu_sched drm_ttm_helper msr drm_exec cdc_acm drm_suballoc_helper snd_sof_pci_intel_tgl snd_sof_pci_intel_cnl snd_sof_intel_hda_generic soundwire_intel soundwire_cadence snd_sof_intel_hda_common snd_soc_hdac_hda snd_sof_intel_hda_mlink snd_sof_intel_hda snd_hda_codec_hdmi snd_sof_pci snd_sof_xtensa_dsp snd_sof snd_sof_utils snd_soc_acpi_intel_match snd_soc_acpi_intel_sdca_quirks intel_rapl_msr soundwire_generic_allocation snd_soc_acpi intel_rapl_common soundwire_bus snd_soc_sdca snd_soc_avs snd_soc_hda_codec snd_hda_ext_core x86_pkg_temp_thermal intel_powerclamp snd_soc_core coretemp snd_compress ac97_bus cmdlinepart snd_pcm_dmaengine kvm_intel spi_nor snd_usb_audio snd_hda_intel snd_intel_dspcfg snd_usbmidi_lib snd_intel_sdw_acpi mtd mei_hdcp mei_pxp snd_hda_codec kvm snd_ump snd_rawmidi snd_hda_core snd_hwdep snd_seq_device snd_pcm irqbypass i2c_i801 rapl snd_timer wmi_bmof intel_cstate hailo_pci(OE) i2c_smbus mei_me snd
[  514.793522]  spi_intel_pci i2c_mux soundcore spi_intel mei mc intel_pmc_core igen6_edac nls_iso8859_1 pmt_telemetry pmt_class intel_vsec acpi_pad acpi_tad upboard_fpga input_leds mac_hid binfmt_misc i915 dm_multipath drm_buddy ttm drm_display_helper cec rc_core i2c_algo_bit efi_pstore nfnetlink dmi_sysfs ip_tables x_tables autofs4 btrfs blake2b_generic raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq raid1 raid0 linear uas usb_storage hid_cmedia spi_pxa2xx_platform dw_dmac dw_dmac_core 8250_dw spi_pxa2xx_core mmc_block rpmb_core sdhci_pci polyval_clmulni sdhci_uhs2 polyval_generic ghash_clmulni_intel r8169 sdhci sha256_ssse3 intel_ish_ipc sha1_ssse3 cqhci intel_lpss_pci realtek intel_ishtp intel_lpss idma64 video wmi pinctrl_alderlake hid_generic usbhid hid aesni_intel crypto_simd cryptd
[  514.793596] CPU: 0 UID: 0 PID: 2245 Comm: det Tainted: G        W  OE      6.14.0-29-generic #29~24.04.1-Ubuntu
[  514.793600] Tainted: [W]=WARN, [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
[  514.793602] Hardware name: AAEON UPS-ASL01/UPS-ASL01, BIOS USASAM12 08/14/2024
[  514.793604] RIP: 0010:find_vma+0x5b/0x70
[  514.793609] Code: c7 c2 ff ff ff ff 48 8d 75 f0 e8 10 ac db 00 48 8b 55 f8 65 48 2b 14 25 28 00 00 00 75 10 c9 31 d2 31 f6 31 ff c3 cc cc cc cc <0f> 0b eb cd e8 8c 3f de 00 66 66 2e 0f 1f 84 00 00 00 00 00 90 90
[  514.793612] RSP: 0018:ffff9de1c3eb7ae0 EFLAGS: 00010246
[  514.793614] RAX: 0000000000000000 RBX: ffff8e49da20f600 RCX: 0000000000000000
[  514.793616] RDX: 0000000000000000 RSI: 00007a45b6830000 RDI: ffff8e49c7a6bc80
[  514.793619] RBP: ffff9de1c3eb7af0 R08: 0000000000000000 R09: 0000000000000000
[  514.793620] R10: 0000000000000000 R11: 0000000000000000 R12: 00007a45b6830000
[  514.793623] R13: ffff8e49c1d8b0c8 R14: 0000000000000000 R15: 0000000000000000
[  514.793625] FS:  00007a45bb7b8740(0000) GS:ffff8e4b37a00000(0000) knlGS:0000000000000000
[  514.793627] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  514.793629] CR2: 00007a45b6833000 CR3: 000000011c9de001 CR4: 0000000000f72ef0
[  514.793632] PKRU: 55555554
[  514.793633] Call Trace:
[  514.793635]  <TASK>
[  514.793637]  hailo_vdma_buffer_map+0x1a6/0x860 [hailo_pci]
[  514.793649]  hailo_vdma_buffer_map_ioctl+0xb4/0x260 [hailo_pci]
[  514.793659]  hailo_vdma_ioctl+0x258/0x360 [hailo_pci]
[  514.793670]  hailo_pcie_fops_unlockedioctl+0x18b/0x6f0 [hailo_pci]
[  514.793680]  __x64_sys_ioctl+0xa4/0xe0
[  514.793686]  x64_sys_call+0x131e/0x2650
[  514.793691]  do_syscall_64+0x7e/0x170
[  514.793695]  ? arch_exit_to_user_mode_prepare.isra.0+0x22/0xd0
[  514.793700]  ? syscall_exit_to_user_mode+0x38/0x1d0
[  514.793704]  ? do_syscall_64+0x8a/0x170
[  514.793707]  ? _raw_spin_lock_irqsave+0xe/0x20
[  514.793711]  ? up+0x58/0xa0
[  514.793716]  ? hailo_pcie_fops_unlockedioctl+0x26d/0x6f0 [hailo_pci]
[  514.793725]  ? __handle_mm_fault+0x62c/0x770
[  514.793729]  ? arch_exit_to_user_mode_prepare.isra.0+0x22/0xd0
[  514.793734]  ? syscall_exit_to_user_mode+0x38/0x1d0
[  514.793738]  ? do_syscall_64+0x8a/0x170
[  514.793741]  ? count_memcg_events.constprop.0+0x2a/0x50
[  514.793745]  ? handle_mm_fault+0x1df/0x2d0
[  514.793749]  ? do_user_addr_fault+0x1d2/0x870
[  514.793754]  ? arch_exit_to_user_mode_prepare.isra.0+0x22/0xd0
[  514.793759]  ? irqentry_exit_to_user_mode+0x2d/0x1d0
[  514.793763]  ? irqentry_exit+0x43/0x50
[  514.793767]  ? exc_page_fault+0x96/0x1e0
[  514.793771]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  514.793776] RIP: 0033:0x7a45b9f24ded
[  514.793779] Code: 04 25 28 00 00 00 48 89 45 c8 31 c0 48 8d 45 10 c7 45 b0 10 00 00 00 48 89 45 b8 48 8d 45 d0 48 89 45 c0 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1a 48 8b 45 c8 64 48 2b 04 25 28 00 00 00
[  514.793781] RSP: 002b:00007ffc0f2123f0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[  514.793784] RAX: ffffffffffffffda RBX: 00007ffc0f212700 RCX: 00007a45b9f24ded
[  514.793787] RDX: 00007ffc0f2124a0 RSI: 00000000c0287604 RDI: 0000000000000003
[  514.793789] RBP: 00007ffc0f212440 R08: 0000000000000000 R09: 0000000000000000
[  514.793791] R10: 0000561cfd292360 R11: 0000000000000246 R12: 0000561cfd292310
[  514.793793] R13: 0000000000000000 R14: 0000561cfd292280 R15: 0000561cfd292338
[  514.793797]  </TASK>
[  514.793798] ---[ end trace 0000000000000000 ]---
2

Welcome to the Hailo Community!

We usually add new kernel support with every release of our software. The next version should be released in October.

That HailoRT version is older than Linux kernel 6.14. So, our developers could not have validated that kernel version.

3

Thank you for your response. That should potentially solve our issue.

Can you confirm, that the new version will be compatible with kernel 6.14 or at least >=6.12 and hailo8, please?

Best regards,

Daniel

4

Looks like my colleagues released HailoRT 4.23 already. You can download it from the Developer Zone. GitHub has not been updated yet.

1 Like
5

We tested it on two machines, but with same result.

Is there anything else we can do to fix this?

6

I am seeing the same errors in dmesg with Linux Mint 22.2 kernel 6.14.0-34. AMD Ryzen 7 7735U.

Newly installed Hailo, both via the latest Docker image and via manual install of Hailo drivers and HailoRT.

hailocli run yolov8m.hef produces expected results as does running tappas/detection/detection.sh. (from the docker install) but produces tons of errors in dmesg.

Is this problem something that will be fixed in the near future or should I look at downgrading the kernel?

7

Hey @user338, @cynfab,

I’ve passed this along to our R&D team to take a look!

In the meantime, I’d suggest trying a lower kernel version - it looks like this might be caused by an invalid argument being passed to the IOCTL, which can vary depending on what kernel you’re running.

8

Rolling back the kernel to 6.8.0-87 eliminates the dmesg errors on both the manually installed as well as the docker container. without re-installing anything.

1 Like
9

Does the new release that was just announced address newer kernels. The LTS kernel is fine for now, but it would be nice to be able to use something newer.

10

Hey @cynfab,

We’ve tested the 4.23 release with the 6.12 kernel. If you’re still seeing this issue with the 6.14 kernel and 4.23, could you send me the log file privately?

11

I should have been more clear in my first post, by “Latest Docker image” that was 4.23.

I’ll try installing a 6.12 kernel and get a bunch of logs for you for both it and 6.14. Need to wait till mid week though as I don’t have access to my hardware at the moment.

1 Like
12

Kernel 6.12 is not used in Ubuntu based distros (amd64) You probably tested it for the Raspberry Pi which released the latest version in Oct 2025 with 6.12.

So I won’t be able to test on my amd64 based Linux Mint 22.2 distro.

I can send you some logs for the 6.14 kernel, what would you like?

13

Hey @cynfab,

Yes , Please!

Send the log in the chat , i will update you when we have something or is it going to be fixed in the next version !

14

The entire dmesg is huge, so I’ll paste the important parts:

First is the kernel startup:

[    0.000000] Linux version 6.14.0-36-generic (buildd@lcy02-amd64-067) (x86_64-linux-gnu-gcc-13 (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0, GNU ld (GNU Binutils for Ubuntu) 2.42) #36~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Oct 15 15:45:17 UTC 2 (Ubuntu 6.14.0-36.36~24.04.1-generic 6.14.11)
[    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-6.14.0-36-generic root=UUID=d81ec7d8-d0ed-4360-99f9-bd95ac94f6ed ro quiet splash
[    0.000000] KERNEL supported cpus:
[    0.000000]   Intel GenuineIntel
[    0.000000]   AMD AuthenticAMD
[    0.000000]   Hygon HygonGenuine
[    0.000000]   Centaur CentaurHauls
[    0.000000]   zhaoxin   Shanghai

The next is:

$ dmesg |grep hailo
[    3.513088] hailo_pci: loading out-of-tree module taints kernel.
[    3.513099] hailo_pci: module verification failed: signature and/or required key missing - tainting kernel
[    3.517365] hailo: Init module. driver version 4.23.0
[    3.517654] hailo 0000:05:00.0: Probing on: 1e60:2864...
[    3.517659] hailo 0000:05:00.0: Probing: Allocate memory for device extension, 9072
[    3.517678] hailo 0000:05:00.0: enabling device (0000 -> 0002)
[    3.518848] hailo 0000:05:00.0: Probing: Device enabled
[    3.518940] hailo 0000:05:00.0: Probing: mapped bar 0 - 0000000061b55ea8 16384
[    3.518948] hailo 0000:05:00.0: Probing: mapped bar 2 - 00000000ec370eba 4096
[    3.518955] hailo 0000:05:00.0: Probing: mapped bar 4 - 000000004f03368b 16384
[    3.518968] hailo 0000:05:00.0: Probing: Setting max_desc_page_size to 4096, (page_size=4096)
[    3.519070] hailo 0000:05:00.0: Probing: Enabled 64 bit dma
[    3.519087] hailo 0000:05:00.0: Probing: Using userspace allocated vdma buffers
[    3.519102] hailo 0000:05:00.0: Disabling ASPM L0s 
[    3.519107] hailo 0000:05:00.0: Successfully disabled ASPM L0s 
[    3.520749] hailo 0000:05:00.0: Writing file hailo/hailo8_fw.bin
[    3.566332] hailo 0000:05:00.0: File hailo/hailo8_fw.bin written successfully
[    3.566338] hailo 0000:05:00.0: Writing file hailo/hailo8_board_cfg.bin
[    3.566400] hailo 0000:05:00.0: File hailo/hailo8_board_cfg.bin written successfully
[    3.566405] hailo 0000:05:00.0: Writing file hailo/hailo8_fw_cfg.bin
[    3.566441] hailo 0000:05:00.0: File hailo/hailo8_fw_cfg.bin written successfully
[    3.643895] hailo 0000:05:00.0: NNC Firmware loaded successfully
[    3.643906] hailo 0000:05:00.0: FW loaded, took 123 ms
[    3.655800] hailo 0000:05:00.0: Probing: Added board 1e60-2864, /dev/hailo0

This is the important part when I resume when I resume a venv:

$ ./hailo_ai_sw_suite_docker_run.sh --resume
Resuming an old container 
hailo8_ai_sw_suite_2025-10_container


Welcome to Hailo AI Software Suite Container
To list available commands, please type:

----------------------------------------------------------------------

HailoRT:                                hailortcli -h
Dataflow Compiler:                      hailo -h
Hailo Model Zoo:                        hailomz -h
Run TAPPAS Detection Application:       tappas/detection/detection.sh

----------------------------------------------------------------------


(hailo_virtualenv) hailo@kenm-EQR7:/local/workspace$ hailo -h
[info] No GPU chosen and no suitable GPU found, falling back to CPU.
[info] Current Time: 07:21:32, 11/30/25
[info] CPU: Architecture: x86_64, Model: AMD Ryzen 7 7735U with Radeon Graphics, Number Of Cores: 16, Utilization: 0.1%
[info] Memory: Total: 27GB, Available: 25GB
[info] System info: OS: Linux, Kernel: 6.14.0-36-generic
[info] Hailo DFC Version: 3.33.0
[info] HailoRT Version: 4.23.0
[info] PCIe: 0000:05:00.0: Number Of Lanes: 2, Speed: 8.0 GT/s PCIe
[info] Running `hailo -h`

And when I run a model in non venv:

hailortcli run yolov8m.hef
Running streaming inference (yolov8m.hef):
  Transform data: true
    Type:      auto
    Quantized: true
Network yolov8m/yolov8m: 100% | 232 | FPS: 46.09 | ETA: 00:00:00
> Inference result:
 Network group: yolov8m
    Frames count: 232
    FPS: 46.09
    Send Rate: 453.08 Mbit/s
    Recv Rate: 450.25 Mbit/s

Doing the above either in the venv or not results in the following appearing in dmesg many times:

[422119.968433] ------------[ cut here ]------------
[422119.968439] WARNING: CPU: 2 PID: 15847 at include/linux/rwsem.h:80 find_vma+0x5b/0x70
[422119.968448] Modules linked in: rfcomm xt_conntrack xt_MASQUERADE bridge stp llc xfrm_user xfrm_algo xt_set ip_set nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 xt_addrtype nft_compat nf_tables ccm overlay qrtr cmac algif_hash algif_skcipher af_alg bnep snd_hda_codec_realtek snd_hda_codec_generic snd_hda_scodec_component snd_hda_codec_hdmi snd_hda_intel intel_rapl_msr amd_atl snd_intel_dspcfg amdgpu intel_rapl_common snd_intel_sdw_acpi iwlmvm snd_hda_codec snd_hda_core snd_hwdep snd_pcm edac_mce_amd amdxcp mac80211 gpu_sched snd_seq_midi snd_seq_midi_event drm_panel_backlight_quirks drm_buddy snd_rawmidi drm_ttm_helper kvm_amd ttm binfmt_misc libarc4 snd_seq drm_exec snd_seq_device drm_suballoc_helper kvm snd_timer btusb drm_display_helper btrtl iwlwifi snd btintel cec btbcm rc_core btmtk irqbypass i2c_piix4 nls_iso8859_1 rapl bluetooth cfg80211 k10temp i2c_smbus i2c_algo_bit ccp hailo_pci(OE) soundcore acpi_tad mac_hid sch_fq_codel msr parport_pc ppdev lp parport efi_pstore nfnetlink dmi_sysfs
[422119.968592]  ip_tables x_tables autofs4 btrfs blake2b_generic raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq raid1 raid0 linear dm_mirror dm_region_hash dm_log uas usb_storage polyval_clmulni nvme polyval_generic ghash_clmulni_intel amd_sfh sha256_ssse3 sha1_ssse3 r8169 hid nvme_core realtek nvme_auth video wmi aesni_intel crypto_simd cryptd
[422119.968655] CPU: 2 UID: 1000 PID: 15847 Comm: hailortcli Tainted: G           OE      6.14.0-36-generic #36~24.04.1-Ubuntu
[422119.968660] Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
[422119.968662] Hardware name: AZW EQ/EQ, BIOS EQR32002 03/26/2025
[422119.968665] RIP: 0010:find_vma+0x5b/0x70
[422119.968669] Code: c7 c2 ff ff ff ff 48 8d 75 f0 e8 20 0c dc 00 48 8b 55 f8 65 48 2b 14 25 28 00 00 00 75 10 c9 31 d2 31 f6 31 ff e9 20 a2 a0 ff <0f> 0b eb cd e8 7c 96 de 00 66 66 2e 0f 1f 84 00 00 00 00 00 90 90
[422119.968672] RSP: 0018:ffffce9a8fb4bba8 EFLAGS: 00010246
[422119.968676] RAX: 0000000000000000 RBX: 00007f389163e000 RCX: 0000000000000000
[422119.968679] RDX: 0000000000000000 RSI: 00007f389163e000 RDI: ffff89b389bf4780
[422119.968681] RBP: ffffce9a8fb4bbb8 R08: 0000000000000000 R09: 0000000000000000
[422119.968683] R10: 0000000000000000 R11: 0000000000000000 R12: ffff89b380dc3700
[422119.968685] R13: 0000000000000000 R14: ffff89b3972ed2f8 R15: 0000000000000000
[422119.968687] FS:  00007f3895109840(0000) GS:ffff89b981b00000(0000) knlGS:0000000000000000
[422119.968690] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[422119.968692] CR2: 00007f3894f80b68 CR3: 00000001174aa000 CR4: 0000000000f50ef0
[422119.968695] PKRU: 55555554
[422119.968697] Call Trace:
[422119.968699]  <TASK>
[422119.968705]  hailo_vdma_buffer_map+0x12e/0x840 [hailo_pci]
[422119.968713]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968721]  hailo_vdma_buffer_map_ioctl+0xb4/0x260 [hailo_pci]
[422119.968729]  hailo_vdma_ioctl+0x10a/0x360 [hailo_pci]
[422119.968735]  hailo_pcie_fops_unlockedioctl+0x18b/0x5d0 [hailo_pci]
[422119.968740]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968743]  ? ksys_mmap_pgoff+0x49/0x240
[422119.968747]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968751]  __x64_sys_ioctl+0xa7/0xe0
[422119.968757]  x64_sys_call+0x131e/0x2650
[422119.968763]  do_syscall_64+0x7e/0x170
[422119.968769]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968772]  ? __handle_mm_fault+0x62c/0x770
[422119.968777]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968780]  ? __count_memcg_events+0xd3/0x1a0
[422119.968786]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968789]  ? count_memcg_events.constprop.0+0x2a/0x50
[422119.968793]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968796]  ? handle_mm_fault+0x22e/0x360
[422119.968800]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968803]  ? do_user_addr_fault+0x5d5/0x870
[422119.968808]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968811]  ? arch_exit_to_user_mode_prepare.isra.0+0x22/0x120
[422119.968816]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968819]  ? irqentry_exit_to_user_mode+0x2d/0x1d0
[422119.968824]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968826]  ? irqentry_exit+0x43/0x50
[422119.968829]  ? srso_alias_return_thunk+0x5/0xfbef5
[422119.968832]  ? exc_page_fault+0x96/0x1e0
[422119.968835]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[422119.968840] RIP: 0033:0x7f3894324e1d
[422119.968843] Code: 04 25 28 00 00 00 48 89 45 c8 31 c0 48 8d 45 10 c7 45 b0 10 00 00 00 48 89 45 b8 48 8d 45 d0 48 89 45 c0 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1a 48 8b 45 c8 64 48 2b 04 25 28 00 00 00
[422119.968846] RSP: 002b:00007fffd4aeb7a0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[422119.968850] RAX: ffffffffffffffda RBX: 00007fffd4aebac0 RCX: 00007f3894324e1d
[422119.968852] RDX: 00007fffd4aeb850 RSI: 00000000c0287604 RDI: 0000000000000005
[422119.968854] RBP: 00007fffd4aeb7f0 R08: 0000000000000002 R09: 0000000000000000
[422119.968856] R10: 0000000004000021 R11: 0000000000000246 R12: 000059f3c6acd550
[422119.968858] R13: 00000000003bd000 R14: 0000000000000002 R15: 000059f3c6acd578
[422119.968864]  </TASK>
[422119.968866] ---[ end trace 0000000000000000 ]---

That’s about it, other than when I revert to Kernel 6.8 I don’t get the errors in dmesg.

15

Hi,

this is a warning only.

Kernel 6.14 is very recent and has stricter VMA locking requirements

· The Hailo out-of-tree driver hasn’t been updated to match these kernel API changes

· When hailo_vdma_buffer_map() tries to map DMA buffers, it’s accessing the process’s memory mappings without proper locking

Report to Hailo - This is a driver bug that Hailo needs to fix. They need to use mmap_read_lock()/mmap_read_unlock() around their find_vma() calls.

here is the fix. After recompiling and rmmod… modprobe your dmesg is free from this messages:

diff -ur linux/vdma/memory.c.orig linux/vdma/memory.c

-– linux/vdma/memory.c.orig 2026-01-29 16:36:47.369352351 +0100

+++ linux/vdma/memory.c 2026-01-29 16:37:34.877361788 +0100

@@ -162,7 +162,9 @@

 }

 if (HAILO_DMA_DMABUF_BUFFER != buffer_type) {

  •         mmap_read_lock(current->mm);

   vma = find_vma(current->mm, user_address);

  •         mmap_read_unlock(current->mm);

   if (IS_ENABLED(HAILO_SUPPORT_MMIO_DMA_MAPPING)) {

       if (NULL == vma) {

           dev_err(dev, "no vma for virt_addr/size = 0x%08lx/0x%08zx\\n", user_address, size);
16

Thanks, I’ll patch that and give it a try. I’ve been wondering if Hailo was ever going to respond. Since the H10 has similar vision processing capabilities as the H8, there are still lots of applications that can use the H8. Their drivers should keep abreast of kernel updates.

17

In fact their most recent driver for the H10 has this lock/unlock around “ vma = find_vma(current->mm, user_address);”

18

After patching, I’m getting the same warnings in dmesg.

I re-did the driver removal, re-compile, modprobe at least twice, to be sure I hadn’t messed it up, which has been known to happen.

Any thoughts on further debugging would be appreciated.