[SOLVED] lspci -k cannot open pci devices

lexan · 2025-11-24 16:06:56

Hello, I have a new installation of Arch on a laptop.

When I run lspci -k, I get the following:

# lspci -k
pcilib: Cannot open /sys/bus/pci/devices/0000:b1:00.0/config
lspci: Cannot open /sys/bus/pci/devices/0000:b1:00.0/vendor: No such file or directory

This behavior is intermittent, but usually gives the error message more times than not, and sometimes the b1 devices changes to b2 in the error message.

I can sometimes get output from lspci -k. lspci without the -k *usually* works. In a recent output that worked, the b2:00.0 device was listed as a USB controller Intel Corporation JHL9580 Thunderbolt 5 with the xhci_hcd kernel driver.

I found a similar issue in this forum: https://forum.qubes-os.org/t/thinkpad-p … re/36771/9

How can I troubleshoot what is going on here and fix this issue?

Thanks

Last edited by lexan (2025-11-26 01:41:35)

seth · 2025-11-24 20:25:34

sudo ls -lR /sys/bus/pci/devices/0000:b1:00.0

lexan · 2025-11-24 20:36:09

seth wrote:

sudo ls -lR /sys/bus/pci/devices/0000:b1:00.0

The error moved to b2, but after trying a few times, I got:

# ls -lR /sys/bus/pci/devices/0000:b2:00.0
lrwxrwxrwx 1 root root 0 Nov 24 15:29 /sys/bus/pci/devices/0000:b2:00.0 -> ../../../devices/pci0000:80/0000:80:1b.4/0000:88:00.0/0000:89:02.0/0000:b2:00.0

seth · 2025-11-24 20:44:35

ls -lR /sys/bus/pci/devices/0000:b2:00.0/ # or
ls -lRL /sys/bus/pci/devices/0000:b2:00.0 #or
ls -lRL /sys/bus/pci/devices/pci0000:80/0000:80:1b.4/0000:88:00.0/0000:89:02.0/0000:b2:00.0

lexan · 2025-11-24 23:01:12

seth wrote:

ls -lR /sys/bus/pci/devices/0000:b2:00.0/ # or
ls -lRL /sys/bus/pci/devices/0000:b2:00.0 #or
ls -lRL /sys/bus/pci/devices/pci0000:80/0000:80:1b.4/0000:88:00.0/0000:89:02.0/0000:b2:00.0

The output of the first command is linked below.

ls -lR /sys/bus/pci/devices/0000:b2:00.0/

http://0x0.st/KWLU.txt

seth · 2025-11-25 09:03:19

-rw-r--r-- 1 root root  4096 Nov 24 17:54 config
-r--r--r-- 1 root root  4096 Nov 24 17:54 vendor

cat /sys/bus/pci/devices/0000:b2:00.0/{config,vendor}

In doubt run "lspci -k", and post the error output and your complete system journal for the boot:

sudo journalctl -b | curl -F 'file=@-' 0x0.st

lexan · 2025-11-25 10:06:33

seth wrote:

In doubt run "lspci -k", and post the error output and your complete system journal for the boot:
sudo journalctl -b | curl -F 'file=@-' 0x0.st

I noticed running "lspci -k" right after boot and login, the error message shows b1. When I run "lspci -k" again, I see expected? output. (At least no error message is displayed.) Then when I run "lspci -k" again, the error message is back with b2.

Below is the journalctl log from boot after the above behavior. "# journalctl -b | curl -F 'file=@-' 0x0.st"

https://0x0.st/KWWJ.txt

seth · 2025-11-25 13:28:26

Nov 25 04:52:26 p16g3 kernel: ------------[ cut here ]------------
Nov 25 04:52:26 p16g3 kernel: thunderbolt 0000:8a:00.0: interrupt for TX ring 0 is already enabled
Nov 25 04:52:26 p16g3 kernel: WARNING: CPU: 5 PID: 1108 at drivers/thunderbolt/nhi.c:146 ring_interrupt_active+0x242/0x2d0 [thunderbolt]
Nov 25 04:52:26 p16g3 kernel: Modules linked in: ccm algif_aead des3_ede_x86_64 des_generic libdes algif_skcipher cmac md4 algif_hash af_alg snd_soc_cs35l56_sdw snd_soc_cs35l56 snd_soc_cs35l56_shared snd_soc_cs_amp_lib snd_soc_rt711_sdca snd_soc_wm_adsp cs42l43_sdw regmap_sdw_mbq snd_soc_dmic cs_dsp cs42l43 regmap_sdw snd_hda_codec_nvhdmi vfat fat snd_hda_codec_intelhdmi snd_sof_pci_intel_mtl snd_sof_intel_hda_generic soundwire_intel snd_sof_intel_hda_sdw_bpt snd_sof_intel_hda_common snd_soc_hdac_hda snd_sof_intel_hda_mlink snd_sof_intel_hda snd_hda_codec_hdmi soundwire_cadence snd_sof_pci snd_sof_xtensa_dsp snd_sof mousedev joydev snd_sof_utils snd_hda_ext_core snd_soc_acpi_intel_match snd_soc_acpi_intel_sdca_quirks soundwire_generic_allocation snd_soc_acpi ucsi_acpi intel_uncore_frequency typec_ucsi soundwire_bus intel_uncore_frequency_common typec x86_pkg_temp_thermal snd_soc_sdca intel_powerclamp roles coretemp snd_hda_intel snd_soc_core uvcvideo snd_hda_codec snd_hda_core videobuf2_vmalloc snd_compress iwlmld iTCO_wdt kvm_intel
Nov 25 04:52:26 p16g3 kernel:  uvc snd_intel_dspcfg processor_thermal_device_pci videobuf2_memops snd_intel_sdw_acpi processor_thermal_device ac97_bus intel_pmc_bxt btusb videobuf2_v4l2 hid_sensor_prox processor_thermal_wt_hint spd5118 snd_hwdep btrtl snd_pcm_dmaengine mei_gsc_proxy videobuf2_common iTCO_vendor_support hid_sensor_trigger platform_temperature_control mac80211 think_lmi kvm industrialio_triggered_buffer i2c_i801 btintel igc processor_thermal_rfim nxp_nci_i2c snd_pcm spi_nor videodev irqbypass hid_multitouch mei_me btbcm i2c_smbus processor_thermal_rapl lenovo_wmi_other kfifo_buf libarc4 rapl btmtk hid_sensor_iio_common nxp_nci lenovo_wmi_helpers intel_cstate wacom ptp intel_rapl_msr intel_uncore psmouse lenovo_wmi_capdata01 pcspkr bluetooth iwlwifi firmware_attributes_class industrialio wmi_bmof mtd mc pps_core intel_rapl_common nci i2c_mux crc8 intel_pmc_core i2c_hid_acpi mei processor_thermal_wt_req snd_timer intel_vpu nfc processor_thermal_power_floor pmt_telemetry i2c_hid processor_thermal_mbox pmt_discovery
Nov 25 04:52:26 p16g3 kernel:  int3403_thermal thunderbolt pmt_class int340x_thermal_zone acpi_tad int3400_thermal pinctrl_meteorpoint intel_hid pinctrl_meteorlake intel_pmc_ssram_telemetry acpi_pad acpi_thermal_rel mac_hid cfg80211 pkcs8_key_parser i2c_dev crypto_user loop nfnetlink ip_tables x_tables dm_crypt encrypted_keys trusted asn1_encoder tee xe drm_suballoc_helper drm_gpusvm_helper nouveau dm_mod mxm_wmi drm_ttm_helper thinkpad_acpi sdhci_pci gpu_sched i915 sdhci_uhs2 drm_gpuvm sparse_keymap platform_profile drm_exec sdhci nvme rfkill drm_buddy i2c_algo_bit cqhci polyval_clmulni ghash_clmulni_intel nvme_core snd ttm intel_gtt intel_lpss_pci spi_intel_pci aesni_intel mmc_core intel_lpss soundcore nvme_keyring spi_intel intel_vsec nvme_auth idma64 drm_display_helper video cec wmi serio_raw hid_sensor_hub
Nov 25 04:52:26 p16g3 kernel: CPU: 5 UID: 0 PID: 1108 Comm: lspci Tainted: G        W           6.17.8-arch1-1 #1 PREEMPT(full)  b229cb54977b6624cce826e2cbd0d8e703fe3921
Nov 25 04:52:26 p16g3 kernel: Tainted: [W]=WARN
Nov 25 04:52:26 p16g3 kernel: Hardware name: LENOVO 21RQCTO1WW/21RQCTO1WW, BIOS N4FET30W (1.11 ) 10/03/2025
Nov 25 04:52:26 p16g3 kernel: RIP: 0010:ring_interrupt_active+0x242/0x2d0 [thunderbolt]
Nov 25 04:52:26 p16g3 kernel: Code: 08 44 89 04 24 e8 ce 6d ce da 44 8b 04 24 48 8b 4c 24 08 4d 89 f1 48 8b 54 24 10 48 89 c6 48 c7 c7 d0 76 63 c1 e8 ee 3d 18 da <0f> 0b 4d 8b 77 08 44 8b 5c 24 04 49 8b 76 18 45 84 e4 0f 85 e9 fe
Nov 25 04:52:26 p16g3 kernel: RSP: 0018:ffffca7820da7760 EFLAGS: 00010046
Nov 25 04:52:26 p16g3 kernel: RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000027
Nov 25 04:52:26 p16g3 kernel: RDX: ffff893f3ff5cfc8 RSI: 0000000000000001 RDI: ffff893f3ff5cfc0
Nov 25 04:52:26 p16g3 kernel: RBP: 0000000000000001 R08: 0000000000000000 R09: 00000000ffffdfff
Nov 25 04:52:26 p16g3 kernel: R10: ffffffff9d59a8a0 R11: ffffca7820da75f8 R12: 0000000000000001
Nov 25 04:52:26 p16g3 kernel: R13: 0000000000038200 R14: ffffffffc16359f7 R15: ffff893bc1e43380
Nov 25 04:52:26 p16g3 kernel: FS:  00007f8fb4ab7780(0000) GS:ffff893fa2a49000(0000) knlGS:0000000000000000
Nov 25 04:52:26 p16g3 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Nov 25 04:52:26 p16g3 kernel: CR2: 000055b56abd0018 CR3: 00000001832a9001 CR4: 0000000000f72ef0
Nov 25 04:52:26 p16g3 kernel: PKRU: 55555554
Nov 25 04:52:26 p16g3 kernel: Call Trace:
Nov 25 04:52:26 p16g3 kernel:  <TASK>
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_pm_runtime_resume+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  tb_ring_start+0x149/0x300 [thunderbolt 0f8827f821735fc0fff5acb6706a18193cd42113]
Nov 25 04:52:26 p16g3 kernel:  ? pci_pme_active+0x1ad/0x1c0
Nov 25 04:52:26 p16g3 kernel:  tb_ctl_start+0x26/0xa0 [thunderbolt 0f8827f821735fc0fff5acb6706a18193cd42113]
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_pm_runtime_resume+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  tb_domain_runtime_resume+0x19/0x40 [thunderbolt 0f8827f821735fc0fff5acb6706a18193cd42113]
Nov 25 04:52:26 p16g3 kernel:  __rpm_callback+0x45/0x1f0
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_pm_runtime_resume+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  rpm_callback+0x6d/0x80
Nov 25 04:52:26 p16g3 kernel:  rpm_resume+0x4af/0x6d0
Nov 25 04:52:26 p16g3 kernel:  ? klist_next+0x193/0x1a0
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_resume_one+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_pm_runtime_resume+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  rpm_get_suppliers+0x4f/0x130
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_pm_runtime_resume+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_pm_runtime_resume+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  __rpm_callback+0x142/0x1f0
Nov 25 04:52:26 p16g3 kernel:  ? __pfx_pci_pm_runtime_resume+0x10/0x10
Nov 25 04:52:26 p16g3 kernel:  rpm_callback+0x6d/0x80
Nov 25 04:52:26 p16g3 kernel:  rpm_resume+0x4af/0x6d0
Nov 25 04:52:26 p16g3 kernel:  ? kernfs_fop_open+0x31e/0x3a0
Nov 25 04:52:26 p16g3 kernel:  ? kernfs_fop_open+0x133/0x3a0
Nov 25 04:52:26 p16g3 kernel:  pm_runtime_barrier+0x87/0x90
Nov 25 04:52:26 p16g3 kernel:  pci_config_pm_runtime_get+0x3a/0x60
Nov 25 04:52:26 p16g3 kernel:  pci_read_config+0x90/0x2a0
Nov 25 04:52:26 p16g3 kernel:  kernfs_fop_read_iter+0xb7/0x1c0
Nov 25 04:52:26 p16g3 kernel:  vfs_read+0x265/0x390
Nov 25 04:52:26 p16g3 kernel:  __x64_sys_pread64+0x9c/0xd0
Nov 25 04:52:26 p16g3 kernel:  do_syscall_64+0x81/0x970
Nov 25 04:52:26 p16g3 kernel:  ? alloc_fd+0x12e/0x190
Nov 25 04:52:26 p16g3 kernel:  ? do_sys_openat2+0xa2/0xe0
Nov 25 04:52:26 p16g3 kernel:  ? set_pte_range+0xf2/0x270
Nov 25 04:52:26 p16g3 kernel:  ? next_uptodate_folio+0x89/0x2a0
Nov 25 04:52:26 p16g3 kernel:  ? filemap_map_pages+0x413/0x680
Nov 25 04:52:26 p16g3 kernel:  ? do_syscall_64+0x81/0x970
Nov 25 04:52:26 p16g3 kernel:  ? do_fault+0x34c/0x5a0
Nov 25 04:52:26 p16g3 kernel:  ? __handle_mm_fault+0x8d1/0xf10
Nov 25 04:52:26 p16g3 kernel:  ? touch_atime+0x1e/0x120
Nov 25 04:52:26 p16g3 kernel:  ? iterate_dir+0x140/0x270
Nov 25 04:52:26 p16g3 kernel:  ? count_memcg_events+0xc2/0x190
Nov 25 04:52:26 p16g3 kernel:  ? handle_mm_fault+0x1d7/0x2d0
Nov 25 04:52:26 p16g3 kernel:  ? do_user_addr_fault+0x21a/0x690
Nov 25 04:52:26 p16g3 kernel:  ? exc_page_fault+0x7e/0x1a0
Nov 25 04:52:26 p16g3 kernel:  entry_SYSCALL_64_after_hwframe+0x76/0x7e
Nov 25 04:52:26 p16g3 kernel: RIP: 0033:0x7f8fb4c9318e
Nov 25 04:52:26 p16g3 kernel: Code: 4d 89 d8 e8 64 be 00 00 4c 8b 5d f8 41 8b 93 08 03 00 00 59 5e 48 83 f8 fc 74 11 c9 c3 0f 1f 80 00 00 00 00 48 8b 45 10 0f 05 <c9> c3 83 e2 39 83 fa 08 75 e7 e8 13 ff ff ff 0f 1f 00 f3 0f 1e fa
Nov 25 04:52:26 p16g3 kernel: RSP: 002b:00007ffd568ed940 EFLAGS: 00000202 ORIG_RAX: 0000000000000011
Nov 25 04:52:26 p16g3 kernel: RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f8fb4c9318e
Nov 25 04:52:26 p16g3 kernel: RDX: 0000000000000040 RSI: 0000555e810fa6b0 RDI: 0000000000000003
Nov 25 04:52:26 p16g3 kernel: RBP: 00007ffd568ed950 R08: 0000000000000000 R09: 0000000000000000
Nov 25 04:52:26 p16g3 kernel: R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000040
Nov 25 04:52:26 p16g3 kernel: R13: 0000555e811073a0 R14: 0000555e810fa6b0 R15: 0000555e69fa6818
Nov 25 04:52:26 p16g3 kernel:  </TASK>
Nov 25 04:52:26 p16g3 kernel: ---[ end trace 0000000000000000 ]---

Nov 25 04:52:26 p16g3 kernel: thunderbolt 0000:8a:00.0: interrupt for TX ring 0 is already enabled
Nov 25 04:52:26 p16g3 kernel: thunderbolt 0000:8a:00.0: interrupt for RX ring 0 is already enabled
Nov 25 04:52:26 p16g3 kernel: thunderbolt 0000:8a:00.0: 0: timeout writing config space 2 to 0x1
Nov 25 04:52:27 p16g3 kernel: thunderbolt 0000:8a:00.0: 0: timeout reading config space 2 from 0x6e
Nov 25 04:52:27 p16g3 kernel: thunderbolt 0000:8a:00.0: 0: timeout reading config space 2 from 0x6b
Nov 25 04:52:27 p16g3 kernel: thunderbolt 0000:8a:00.0: 0: failed to restore TMU configuration
Nov 25 04:52:28 p16g3 kernel: thunderbolt 0000:8a:00.0: 0: timeout writing config space 2 to 0x19
Nov 25 04:52:28 p16g3 kernel: thunderbolt 0000:8a:00.0: 0: timeout writing config space 2 to 0x19

Do the

pcie_aspm=off pcie_port_pm=off

https://wiki.archlinux.org/title/Kernel_parameters change anything about that?

lexan · 2025-11-25 16:14:09

seth wrote:

Do the
pcie_aspm=off pcie_port_pm=off
https://wiki.archlinux.org/title/Kernel_parameters change anything about that?

Yes: Both the "WARNING" blocks for the thunderbolt sections as well as those "interrupt," "timeout," and "failed" messages are gone.

seth · 2025-11-25 18:44:44

Does it also impact the lspci behavior?
You could try which parameter is critical. Right now it looks like the thunderbolt controller is just very slow to wake up.
Does this cause real-life problems? Otherwise I'd frankly ignore it rather than disabling power saving.
There might also be UEFI/BIOS options (in doubt to deactivate thunderbolt if you don't use it)

lexan · 2025-11-25 19:25:32

seth wrote:

Does it also impact the lspci behavior?
You could try which parameter is critical. Right now it looks like the thunderbolt controller is just very slow to wake up.
Does this cause real-life problems? Otherwise I'd frankly ignore it rather than disabling power saving.
There might also be UEFI/BIOS options (in doubt to deactivate thunderbolt if you don't use it)

The lspci behavior seems to be fixed when disabling pci aspm and port pm. I haven't noticed it since. I made a udev rule as suggested in the ASPM section of the Power management wiki:

# blacklist for pci runtime power management
SUBSYSTEM=="pci", ATTR{vendor}=="0x8086", ATTR{device}=="0x5781", ATTR{power/control}="on"
SUBSYSTEM=="pci", ATTR{vendor}=="0x8086", ATTR{device}=="0x5782", ATTR{power/control}="on", GOTO="pci_pm_end"

SUBSYSTEM=="pci", ATTR{power/control}="auto"
LABEL="pci_pm_end"

As for real-world problems, I have a sneaky suspicion that something about the ASPM is preventing the system from fully/cleanly powering off. (i.e. It hangs after reaching target System Power Off, shuts down, and stops the journal... yet power to the laptop is still there and I have to hold the power button to actually completely power it off.) I found similar behavior here: Linux stuck/frozen after shutdown

Jul 18 03:41:49 pc. systemd[1]: systemd-poweroff.service: Deactivated successfully.
Jul 18 03:41:49 pc. systemd[1]: Finished System Power Off.
Jul 18 03:41:49 pc. systemd[1]: Reached target System Power Off.
Jul 18 03:41:49 pc. systemd[1]: Shutting down.
Jul 18 03:41:49 pc. systemd-shutdown[1]: Syncing filesystems and block devices.
Jul 18 03:41:49 pc. systemd-shutdown[1]: Sending SIGTERM to remaining processes...
Jul 18 03:41:49 pc. systemd-journald[281]: Received SIGTERM from PID 1 (systemd-shutdow).
Jul 18 03:41:49 pc. systemd-journald[281]: Journal stopped

Since disabling ASPM, I haven't noticed the laptop hang after trying to shutdown.

It would be nice to keep the power saving, because out-of-the-box with no optimization, the computer idles around 4 Watts, but with the kernel parameters described above, it idles around 12 Watts.
Maybe there's a way to wake the thunderbolt controller up before shutdown, or introduce a small delay?

I *do* notice a slight lag in the output of "lspci -k" before it prints the output, suggesting that it’s waking up -- and then when I run "lspci -k" quickly after, the output is more immediate.

Last edited by lexan (2025-11-25 19:54:16)

seth · 2025-11-25 20:56:22

Maybe there's a way to wake the thunderbolt controller up before shutdown, or introduce a small delay?

https://wiki.archlinux.org/title/Udisks … t_shutdown
But instead just run /usr/bin/lspci -k"?

lexan · 2025-11-26 01:40:57

seth wrote:

Maybe there's a way to wake the thunderbolt controller up before shutdown, or introduce a small delay?
https://wiki.archlinux.org/title/Udisks … t_shutdown
But instead just run /usr/bin/lspci -k"?

Thanks a lot, seth. Yes, something like that will work! I've been running 'lspci' to wake up the bus before sending the poweroff signal, and I haven't had an issue since.

lexan · 2025-11-27 16:16:21

Script to run before shutdown placed in "/usr/local/bin/execute-lspci.sh":

#!/bin/bash -u
# place in `/usr/local/bin/execute-lspci.sh`
# make executable `chmod 0755 execute-lspci.sh`

/usr/bin/lspci > /dev/null 2>&1

Oneshot systemd service in "/etc/systemd/system/execute-lspci-before-shutdown.service":

[Unit]
Description=Execute lspci script before system poweroff
DefaultDependencies=no
Before=shutdown.target

[Service]
Type=oneshot
ExecStart=/usr/local/bin/execute-lspci.sh

[Install]
WantedBy=shutdown.target

Then reload the daemons with `sudo systemctl daemon-reload` and enable/start the new service with `sudo systemctl enable execute-lspci-before-shutdown.service`

Arch Linux

#1 2025-11-24 16:06:56

[SOLVED] lspci -k cannot open pci devices

#2 2025-11-24 20:25:34

Re: [SOLVED] lspci -k cannot open pci devices

#3 2025-11-24 20:36:09

Re: [SOLVED] lspci -k cannot open pci devices

#4 2025-11-24 20:44:35

Re: [SOLVED] lspci -k cannot open pci devices

#5 2025-11-24 23:01:12

Re: [SOLVED] lspci -k cannot open pci devices

#6 2025-11-25 09:03:19

Re: [SOLVED] lspci -k cannot open pci devices

#7 2025-11-25 10:06:33

Re: [SOLVED] lspci -k cannot open pci devices

#8 2025-11-25 13:28:26

Re: [SOLVED] lspci -k cannot open pci devices

#9 2025-11-25 16:14:09

Re: [SOLVED] lspci -k cannot open pci devices

#10 2025-11-25 18:44:44

Re: [SOLVED] lspci -k cannot open pci devices

#11 2025-11-25 19:25:32

Re: [SOLVED] lspci -k cannot open pci devices

#12 2025-11-25 20:56:22

Re: [SOLVED] lspci -k cannot open pci devices

#13 2025-11-26 01:40:57

Re: [SOLVED] lspci -k cannot open pci devices

#14 2025-11-27 16:16:21

Re: [SOLVED] lspci -k cannot open pci devices

Board footer