irq/9-acpi interrupts and high cpu usage

doums · 2025-12-14 01:44:54

Hi, I have this process with a relatively high cpu footprint, in the top ~3 processes when system is in idle.
I've been observing it since day one on a new recent laptop (ThinkPad X1 carbon gen 13, Lunar Lake cpu).
It is constantly generating a lot of acpi interrupts, indefinitely until system shutdown

cat /sys/firmware/acpi/interrupts/sci /sys/firmware/acpi/interrupts/gpe6E
   22458
   22457  EN     enabled      unmasked

cat /proc/interrupts | rg '\b9:'
   9:      24447          0          0          0          0          0          0          0 IR-IO-APIC    9-fasteoi   acpi

(this after ~2min, right after a fresh boot)

Thankfully, there is no other symptom other than the cpu usage.

If I mask it with `acpi_mask_gpe=0x6E` then it fixes the interrupts and cpu. But it introduces weird lags when changing the monitor brightness, like when I press the "brightness Down or Up" keys the backlight change will occur after some delay like 1.5 second, for each step.
I had to revert it.

Here are some kernel logs (I can share the whole ring buffer, just letme know I will send via pm)

--
[    0.034801] On node 0, zone Normal: 25856 pages in unavailable ranges
[    0.034809] On node 0, zone Normal: 2048 pages in unavailable ranges
[    0.035307] ACPI: PM-Timer IO Port: 0x1808
[    0.035312] ACPI: X2APIC_NMI (uid[0xffffffff] high level lint[0x1])
[    0.035352] IOAPIC[0]: apic_id 2, version 32, address 0xfec00000, GSI 0-119
[    0.035354] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
[    0.035355] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level)
[    0.035357] ACPI: Using ACPI (MADT) for SMP configuration information
[    0.035358] ACPI: HPET id: 0x8086a701 base: 0xfed00000
[    0.035366] e820: update [mem 0x7ce98000-0x7cf73fff] usable ==> reserved
[    0.035372] TSC deadline timer available
--
[    0.088353] RCU Tasks Rude: Setting shift to 3 and lim to 1 rcu_task_cb_adjust=1 rcu_task_cpu_ids=8.
[    0.088353] RCU Tasks Trace: Setting shift to 3 and lim to 1 rcu_task_cb_adjust=1 rcu_task_cpu_ids=8.
[    0.090733] NR_IRQS: 524544, nr_irqs: 2048, preallocated irqs: 16
[    0.091009] rcu: srcu_init: Setting srcu_struct sizes based on contention.
[    0.091258] kfence: initialized - using 2097152 bytes for 255 objects at 0x(____ptrval____)-0x(____ptrval____)
[    0.091279] Console: colour dummy device 80x25
[    0.091281] printk: legacy console [tty0] enabled
[    0.091307] ACPI: Core revision 20250404
[    0.091493] hpet: HPET dysfunctional in PC10. Force disabled.
[    0.091590] APIC: Switch to symmetric I/O mode setup
--
[    0.091624] DMAR-IR: HPET id 0 under DRHD base 0xfc820000
[    0.091624] DMAR-IR: Queued invalidation will be enabled to support x2apic and Intr-remapping.
[    0.093430] DMAR-IR: Enabled IRQ remapping in x2apic mode
[    0.097555] clocksource: tsc-early: mask: 0xffffffffffffffff max_cycles: 0x2f9a2767029, max_idle_ns: 440795219692 ns
[    0.097561] Calibrating delay loop (skipped), value calculated using timer frequency.. 6604.80 BogoMIPS (lpj=3302400)
--
[    0.109651] devtmpfs: initialized
[    0.109651] x86/mm: Memory block size: 128MB
[    0.111000] ACPI: PM: Registering ACPI NVS region [mem 0x8d32f000-0x8df2efff] (12582912 bytes)
[    0.111000] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 1911260446275000 ns
[    0.111000] posixtimers hash table entries: 4096 (order: 4, 65536 bytes, linear)
--
[    0.112622] cpuidle: using governor ladder
[    0.112624] cpuidle: using governor menu
[    0.112643] acpiphp: ACPI Hot Plug PCI Controller Driver version: 0.5
[    0.112849] PCI: ECAM [mem 0xc0000000-0xcfffffff] (base 0xc0000000) for domain 0000 [bus 00-ff]
[    0.112859] PCI: Using configuration type 1 for base access
--
[    0.113827] raid6: skipped pq benchmark and selected avx2x4
[    0.113828] raid6: using avx2x2 recovery algorithm
[    0.113867] ACPI: Added _OSI(Module Device)
[    0.113868] ACPI: Added _OSI(Processor Device)
[    0.113868] ACPI: Added _OSI(Processor Aggregator Device)
[    0.140743] ACPI: 27 ACPI AML tables successfully acquired and loaded
[    0.141399] ACPI: EC: EC started
[    0.141399] ACPI: EC: interrupt blocked
[    0.141771] ACPI: EC: EC_CMD/EC_SC=0x66, EC_DATA=0x62
[    0.141782] ACPI: EC: Boot ECDT EC used to handle transactions
[    0.146056] ACPI: [Firmware Bug]: BIOS _OSI(Linux) query ignored
[    0.183581] ACPI: USB4 _OSC: OS supports USB3+ DisplayPort+ PCIe+ XDomain+
[    0.183585] ACPI: USB4 _OSC: OS controls USB3+ DisplayPort+ PCIe+ XDomain+
[    0.186536] ACPI: Interpreter enabled
[    0.186624] ACPI: PM: (supports S0 S4 S5)
[    0.186627] ACPI: Using IOAPIC for interrupt routing
[    0.188756] PCI: Using host bridge windows from ACPI; if necessary, use "pci=nocrs" and report a bug
[    0.188758] PCI: Ignoring E820 reservations for host bridge windows
[    0.189747] ACPI: Enabled 10 GPEs in block 00 to 7F
[    0.189800] ACPI: Enabled 13 GPEs in block 80 to DF
[    0.200206] ACPI: \_SB_.PC00.LPCB.EC__.PUBS: New power resource
[    0.203861] ACPI: \_SB_.PC00.XHCI.RHUB.HS05.PRT3.WWPR: New power resource
[    0.204046] ACPI: \_SB_.PC00.XHCI.RHUB.HS06.BTRT: New power resource
[    0.204124] ACPI: \_SB_.PC00.XHCI.RHUB.HS06.DBTR: New power resource
[    0.219043] ACPI: \_SB_.PC00.CNVW.WRST: New power resource
[    0.219286] ACPI: \_SB_.PC00.BTPC.WRST: New power resource
[    0.220942] ACPI: \_SB_.PC00.RP02.PXP_: New power resource
[    0.222360] ACPI: \_SB_.PC00.RP03.PXSX.WRST: New power resource
[    0.222677] ACPI: \_SB_.PC00.RP04.PXP_: New power resource
[    0.223587] ACPI: \_SB_.PC00.RP04.PXSX.MRST: New power resource
[    0.225686] ACPI: \_SB_.PC00.TBT0: New power resource
[    0.225715] ACPI: \_SB_.PC00.TBT1: New power resource
[    0.225741] ACPI: \_SB_.PC00.D3C_: New power resource
[    0.247880] ACPI: \PIN_: New power resource
[    0.247906] ACPI: \PINP: New power resource
[    0.248789] ACPI: PCI Root Bridge [PC00] (domain 0000 [bus 00-fe])
[    0.248801] acpi PNP0A08:00: _OSC: OS supports [ExtendedConfig ASPM ClockPM Segments MSI EDR HPX-Type3]
[    0.251259] acpi PNP0A08:00: _OSC: OS now controls [PCIeHotplug SHPCHotplug PME AER PCIeCapability LTR DPC]
[    0.253710] PCI host bridge to bus 0000:00
[    0.253716] pci_bus 0000:00: root bus resource [io  0x0000-0x0cf7 window]
--
[    0.291144] pci 0000:00:1c.4: ASPM: current common clock configuration is inconsistent, reconfiguring
[    0.291202] pci 0000:00:1c.4: PCI bridge to [bus 04]
[    0.297107] ACPI: \_SB_.PEPD: Duplicate LPS0 _DSM functions (mask: 0x1)
[    0.792529] Low-power S0 idle used by default for system suspend
[    0.793837] ACPI: EC: interrupt unblocked
[    0.793838] ACPI: EC: event unblocked
[    0.793859] ACPI: EC: EC_CMD/EC_SC=0x66, EC_DATA=0x62
[    0.793860] ACPI: EC: GPE=0x6e
[    0.793862] ACPI: \_SB_.PC00.LPCB.EC__: Boot ECDT EC initialization complete
[    0.793863] ACPI: \_SB_.PC00.LPCB.EC__: EC: Used to handle transactions and events
[    0.794010] iommu: Default domain type: Translated
[    0.794010] iommu: DMA domain TLB invalidation policy: lazy mode
[    0.794753] SCSI subsystem initialized
[    0.794766] libata version 3.00 loaded.
[    0.794766] ACPI: bus type USB registered
[    0.794766] usbcore: registered new interface driver usbfs
[    0.794766] usbcore: registered new interface driver hub
--
[    0.796075] mctp: management component transport protocol core
[    0.796075] NET: Registered PF_MCTP protocol family
[    0.796075] PCI: Using ACPI for IRQ routing
[    0.814262] PCI: pci_cache_line_size set to 64 bytes
[    0.814584] pci 0000:00:1f.5: BAR 0 [mem 0xfe010000-0xfe010fff]: can't claim; no compatible bridge window
--
[    0.816871] VFS: Disk quotas dquot_6.6.0
[    0.816871] VFS: Dquot-cache hash table entries: 512 (order 0, 4096 bytes)
[    0.817012] pnp: PnP ACPI init
[    0.818545] system 00:00: [io  0x0680-0x069f] has been reserved
[    0.818550] system 00:00: [io  0x164e-0x164f] has been reserved
--
[    0.821101] system 00:08: [mem 0xfec00000-0xfed3ffff] could not be reserved
[    0.821103] system 00:08: [mem 0xfed4c000-0xffffffff] could not be reserved
[    0.821309] pnp: PnP ACPI: found 9 devices
[    0.827060] clocksource: acpi_pm: mask: 0xffffff max_cycles: 0xffffff, max_idle_ns: 2085701024 ns
[    0.827139] NET: Registered PF_INET protocol family
[    0.827398] IP idents hash table entries: 262144 (order: 9, 2097152 bytes, linear)
--
[    0.900293] io scheduler bfq registered
[    0.900925] ledtrig-cpu: registered to indicate activity on CPUs
[    0.901205] pcieport 0000:00:07.0: PME: Signaling with IRQ 126
[    0.901323] pcieport 0000:00:07.0: AER: enabled with IRQ 126
[    0.901352] pcieport 0000:00:07.0: pciehp: Slot #6 AttnBtn- PwrCtrl- MRL- AttnInd- PwrInd- HotPlug+ Surprise+ Interlock- NoCompl+ IbPresDis- LLActRep+
[    0.901817] pcieport 0000:00:07.2: PME: Signaling with IRQ 127
[    0.901907] pcieport 0000:00:07.2: AER: enabled with IRQ 127
[    0.901930] pcieport 0000:00:07.2: pciehp: Slot #8 AttnBtn- PwrCtrl- MRL- AttnInd- PwrInd- HotPlug+ Surprise+ Interlock- NoCompl+ IbPresDis- LLActRep+
[    0.902526] pcieport 0000:00:1c.0: PME: Signaling with IRQ 128
[    0.902715] pcieport 0000:00:1c.0: pciehp: Slot #0 AttnBtn- PwrCtrl- MRL- AttnInd- PwrInd- HotPlug+ Surprise+ Interlock- NoCompl+ IbPresDis- LLActRep+
[    0.903620] pcieport 0000:00:1c.4: PME: Signaling with IRQ 129
[    0.903868] pcieport 0000:00:1c.4: AER: enabled with IRQ 129
[    0.904916] ACPI: AC: AC Adapter [AC] (on-line)
[    0.904978] input: Sleep Button as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0C0E:00/input/input0
[    0.905135] Freeing initrd memory: 19712K
[    0.909513] ACPI: button: Sleep Button [SLPB]
[    0.909662] input: Lid Switch as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0C0D:00/input/input1
[    0.909805] ACPI: button: Lid Switch [LID]
[    0.909919] input: Power Button as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0C0C:00/input/input2
[    0.909997] ACPI: button: Power Button [PWRB]
[    0.924384] thermal LNXTHERM:00: registered as thermal_zone0
[    0.924395] ACPI: thermal: Thermal Zone [THM0] (49 C)
[    0.925317] Serial: 8250/16550 driver, 32 ports, IRQ sharing enabled
[    0.933322] hpet_acpi_add: no address or irqs in _CRS
[    0.933508] Non-volatile memory driver v1.3
[    0.933514] Linux agpgart interface v0.103
[    0.935548] tpm_tis NTC0702:00: 2.0 TPM (device-id 0xFC, rev-id 1)
[    0.972254] ACPI: battery: Slot [BAT0] (battery present)
[    1.051336] ACPI: bus type drm_connector registered
[    1.054604] xhci_hcd 0000:00:0d.0: xHCI Host Controller
[    1.054631] xhci_hcd 0000:00:0d.0: new USB bus registered, assigned bus number 1
--
[    1.076504] usbcore: registered new interface driver usbserial_generic
[    1.076532] usbserial: USB Serial support registered for generic
[    1.076686] i8042: PNP: PS/2 Controller [PNP0303:KBD,PNP0f13:MOU] at 0x60,0x64 irq 1,12
[    1.077103] i8042: Warning: Keylock active
[    1.078884] serio: i8042 KBD port at 0x60,0x64 irq 1
[    1.079072] serio: i8042 AUX port at 0x60,0x64 irq 12
[    1.079648] rtc_cmos rtc_cmos: RTC can wake from S4
[    1.081163] input: AT Translated Set 2 keyboard as /devices/platform/i8042/serio0/input/input3
--
[    1.353422] idma64 idma64.1: Found Intel integrated DMA 64-bit
[    1.360665] Key type psk registered
[    1.376985] thinkpad_acpi: ThinkPad ACPI Extras v0.26
[    1.376989] thinkpad_acpi: http://ibm-acpi.sf.net/
[    1.376990] thinkpad_acpi: ThinkPad BIOS N4BET70W (1.40 ), EC N4BHT57W
[    1.376991] thinkpad_acpi: Lenovo ThinkPad X1 Carbon Gen 13, model 21NSCTO1WW
[    1.377253] thinkpad_acpi: radio switch found; radios are enabled
[    1.377454] thinkpad_acpi: This ThinkPad has standard ACPI backlight brightness control, supported by the ACPI video driver
[    1.377455] thinkpad_acpi: Disabling thinkpad-acpi brightness events by default...
[    1.378217] nvme 0000:04:00.0: platform quirk: setting simple suspend
[    1.378289] nvme nvme0: pci function 0000:04:00.0
[    1.379791] thinkpad_acpi: rfkill switch tpacpi_bluetooth_sw: radio is unblocked
[    1.383683] nvme nvme0: D3 entry latency set to 10 seconds
[    1.386533] nvme nvme0: 8/0/0 default/read/poll queues
[    1.390366]  nvme0n1: p1 p2 p3 p4 p5
[    1.396473] thinkpad_acpi: secondary fan control detected & enabled
[    1.400829] thinkpad_acpi: battery 1 registered (start 0, stop 100, behaviours: 0xb)
[    1.400849] ACPI: battery: new hook: ThinkPad Battery Extension
[    1.407863] input: ThinkPad Extra Buttons as /devices/platform/thinkpad_acpi/input/input5
[    1.440025] usb 3-3: New USB device found, idVendor=046d, idProduct=c547, bcdDevice= 4.02
[    1.440030] usb 3-3: New USB device strings: Mfr=1, Product=2, SerialNumber=0
--
[    2.355735] xe 0000:00:02.0: [drm] Registered 3 planes with drm panic
[    2.355742] [drm] Initialized xe 1.1.0 for 0000:00:02.0 on minor 0
[    2.358698] ACPI: video: Video Device [GFX0] (multi-head: yes  rom: no  post: no)
[    2.359096] input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/LNXVIDEO:00/input/input8
[    2.394755] fbcon: xedrmfb (fb0) is primary device
--
[   17.217455] intel_pmc_core INT33A1:00:  initialized
[   17.231305] intel_vpu 0000:00:0b.0: enabling device (0000 -> 0002)
[   17.231973] ACPI: bus type thunderbolt registered
[   17.236889] intel_vpu 0000:00:0b.0: [drm] Firmware: intel/vpu/vpu_40xx_v1.bin, version: Sep 25 2025*NPU40xx*build/ci/***REDACTED***
[   17.236891] intel_vpu 0000:00:0b.0: [drm] Scheduler mode: HW
--
[   17.596996] sof-audio-pci-intel-lnl 0000:00:1f.3: bound 0000:00:02.0 (ops intel_audio_component_bind_ops [xe])
[   17.604731] sof-audio-pci-intel-lnl 0000:00:1f.3: use msi interrupt mode
[   17.615041] acpi device:2a: find_sdca_function: SDCA function UAJ (type 6) at 0x1
[   17.615209] acpi device:2b: find_sdca_function: SDCA function HID (type 10) at 0x3
[   17.619842] acpi device:2d: find_sdca_function: SDCA function SmartAmp (type 1) at 0x4
[   17.632109] sof-audio-pci-intel-lnl 0000:00:1f.3: hda codecs found, mask 4
[   17.632113] sof-audio-pci-intel-lnl 0000:00:1f.3: NHLT device BT(0) detected, ssp_mask 0x4

Last edited by doums (2025-12-14 01:46:09)

xerxes_ · 2025-12-14 11:26:41

Alter your acpi workaround and run 'watch lsirq' command and post what is doing the most IRQs.

Also how high is your CPU usage by, for example, top and what CPU do you have?

Last edited by xerxes_ (2025-12-14 11:32:40)

doums · 2025-12-14 12:19:34

These are the top of the list watching them live 0.2 interval, they are constantly flowing:

* gpe6E unmasked
(except irq 30 and 138, not moving)

lsirq -t 10000
IRQ    TOTAL NAME
CAL 29160036 Function call interrupts
LOC 23506525 Local timer interrupts
 30  1570625 IR-IO-APIC 30-fasteoi idma64.1, i2c_designware.1
TRM  1519780 Thermal event interrupts
155  1404181 IR-PCI-MSI-0000:00:02.0 0-edge xe
IWI   939624 IRQ work interrupts
TLB   724319 TLB shootdowns
  9   339687 IR-IO-APIC 9-fasteoi acpi
138   196988 IR-PCI-MSI-0000:00:14.0 0-edge xhci_hcd
RES   152178 Rescheduling interrupts
174   130000 IR-PCI-MSIX-0000:00:14.3 0-edge iwlwifi:default_queue
181   107100 IR-PCI-MSIX-0000:00:14.3 7-edge iwlwifi:queue_7
156    88602 intel-gpio 29 SNSL0028:00
 14    88602 IR-IO-APIC 14-fasteoi INTC105D:00, INTC105D:01, INTC105D:02, INTC105D:03, INTC105D:04

* gpe6E masked
all flowing except for irq 30, 201, 14, 168

IRQ  TOTAL NAME
LOC 391685 Local timer interrupts
CAL 338449 Function call interrupts
 30  47158 IR-IO-APIC 30-fasteoi idma64.1, i2c_designware.1
TRM  19884 Thermal event interrupts
155  17384 IR-PCI-MSI-0000:00:02.0 0-edge xe
201   7665 IR-PCI-MSIX-0000:00:14.7 0-edge btintel_pcie
IWI   7604 IRQ work interrupts
RES   7497 Rescheduling interrupts
TLB   6151 TLB shootdowns
168   3499 intel-gpio 29 SNSL0028:00
 14   3499 IR-IO-APIC 14-fasteoi INTC105D:00, INTC105D:01, INTC105D:02, INTC105D:03, INTC105D:04
163   2966 IR-PCI-MSIX-0000:00:14.3 0-edge iwlwifi:default_queue
--
177   1000 IR-PCI-MSIX-0000:00:14.3 6-edge iwlwifi:queue_6

Cpu is intel LunarLake ultra 7 268V, as I said, the footprint is relative i.e. negligible compared to real application workload.
But at idle, after a fresh boot and logged in into the X session for example, ` irq/9-acpi` is at the top of the list in terms of cpu usage at ~0.2-0.4%
EDIT: `top` rate it at ~1.7%

EDIT: not sure it is relevant here, but sharing some kernel msgs

[   10.919578] iwlwifi 0000:00:14.3: enabling device (0000 -> 0002)
[   10.924628] iwlwifi 0000:00:14.3: Detected crf-id 0xbadcafe, cnv-id 0x1080900 wfpm id 0x80005b20
[   10.924645] iwlwifi 0000:00:14.3: PCI dev a840/00e0, rev=0x461, rfid=0x20112200
[   10.924648] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 7 BE201 320MHz
[   10.924709] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-bz-b0-fm-c0-102.ucode failed with error -2
--
[   10.929782] iwlwifi 0000:00:14.3: loaded firmware version 101.6ef20b19.0 bz-b0-fm-c0-101.ucode op_mode iwlmld
--
[   11.532843] iwlwifi 0000:00:14.3: Detected RF FM, rfid=0x20112200
[   11.627990] sof-audio-pci-intel-lnl 0000:00:1f.3: Booted firmware version: 2.13.0.1
[   11.639855] iwlwifi 0000:00:14.3: base HW address: REDACTED

Last edited by doums (2025-12-14 12:35:48)

xerxes_ · 2025-12-14 20:25:06

Worth trying? :
https://forum.manjaro.org/t/fixed-my-ir … ing/181451

If not, then post output of command:

sudo journalctl -b -g 'AE_NOT_FOUND|OSC'

when acpi is not masked by acpi_mask_gpe .

Last edited by xerxes_ (2025-12-14 20:27:36)

doums · 2025-12-14 22:29:10

> just replug all USB-Connections

This is useless in my case, the irq9 issue happens regardless of usb device connected.
ie. it's there when nothing is connected via usb to the laptop.

Dec 14 13:22:10 tarse kernel: ACPI: USB4 _OSC: OS supports USB3+ DisplayPort+ PCIe+ XDomain+
Dec 14 13:22:10 tarse kernel: ACPI: USB4 _OSC: OS controls USB3+ DisplayPort+ PCIe+ XDomain+
Dec 14 13:22:10 tarse kernel: acpi PNP0A08:00: _OSC: OS supports [ExtendedConfig ASPM ClockPM Segments MSI EDR HPX-Type3]
Dec 14 13:22:10 tarse kernel: acpi PNP0A08:00: _OSC: OS now controls [PCIeHotplug SHPCHotplug PME AER PCIeCapability LTR DPC]

Last edited by doums (2025-12-14 22:29:32)

xerxes_ · 2025-12-15 20:57:15

I'm sorry. For now, I don't know what more I can advise you, beside to post full journalctl log (sudo journalctl -b), so maybe we could find something what points someone to some solution.

Your system bios don't return any obvious errors (at least for me) and 1,7% of additional CPU or system load is not so much (unless you are often on battery) or until you find solution with time.

So for now you have to choose what is for you more acceptable: 1,7% of additional CPU load or 1.5 second of backlight change delay.

ULTD · 2026-05-06 07:02:26

Just a partial confirmation: Some weeks ago on my openSUSE system (Linux 6.19.12 on x86 32-bit Thinkpad X60) I had the same problem after wakeup. It has gone away after heavily unplugging/plugging something (USB device or power supply). Later I never noticed a high CPU load from irq/9-acpi. It is consuming CPU time similar to jbd2, udisksd or xfce4-screensaver.

Last edited by ULTD (2026-05-06 08:53:37)

Lone_Wolf · 2026-05-06 10:23:46

Moderator Note
OP didn't reply to previous post and hasn't visited the forum since january.
It looks like they are no longer interested in solving this.

Closing the thread.

Added
Op contacted me through forum mail and asked for reopening. Next time please use the report button so all moderators see the request.

For clarity : The previous post that wasn't replied to referred to post #5 from december.
Re-opening as OP is still interested in finding a solution .

Last edited by Lone_Wolf (2026-05-07 07:08:00)

doums · 2026-05-07 08:55:07

Since post #5 it's statu quo as I understand it.

I tested again yesterday on 7.0.3-arch, unmasked gpe 0x6E and irq/9-acpi is still at the top of the process list, using bottom to monitor.
Continuously producing a lot of interrupts as before it seems.

Last edited by doums (2026-05-07 08:56:28)

xerxes_ · 2026-05-07 16:13:03

Wild try: add to kernel boot options: "acpi=noirq".

doums · 2026-05-07 17:07:01

I've been digging a bit more, and I think since new kernel 7.0* the interrupts rate is definitely lower than before. Sitting around ~10-30 per seconds.
(using `watch -n1 "rg '\b9:' /proc/interrupts"`)
Initially when I reported the issue, the rate was thousands interrupts/sec if I remember correctly.
Also, the cpu usage seems lower too, even if still at the top of the cpu list on idle (bottom, top).

I asked some llm

irq/9-acpi at 0-2% on a ThinkPad X1 Carbon with this level of ACPI firmware complexity (27 AML tables, dual fan control, DPTF thermal management, Thunderbolt, USB-C power resources...) is exactly expected behavior. Nothing to fix.
10-50 interrupts/second is completely normal for an active ThinkPad. That's just ACPI doing its job — thermal polling, EC events, power management ticks.
irq/9-acpi appearing at the top of bottom is purely a sorting artifact — it wins the top spot because it generates small but frequent CPU wakeups, while other processes do more work in longer bursts. It's not actually consuming significant CPU.

Makes sense to me, but I don't have the knowledge to confirm.
If confirmed, then I think we can consider this issue solved.

xerxes_ wrote:

Wild try: add to kernel boot options: "acpi=noirq".

Yeah, could be something to try, although sounds a bit yolo xD kernel docs says "Don’t route interrupts" (not sure of the implications)

EDIT: Let's close this thread then. As I said generated interrupt rate from irq/9-acpi are reduced compared to previous kernels. In any case now we have some traces, if someone want to reopen the topic.

Last edited by doums (2026-05-10 18:35:38)

WorMzy · 2026-05-11 18:23:44

Mod note: Closing, on request.

Arch Linux

#1 2025-12-14 01:44:54

irq/9-acpi interrupts and high cpu usage

#2 2025-12-14 11:26:41

Re: irq/9-acpi interrupts and high cpu usage

#3 2025-12-14 12:19:34

Re: irq/9-acpi interrupts and high cpu usage

#4 2025-12-14 20:25:06

Re: irq/9-acpi interrupts and high cpu usage

#5 2025-12-14 22:29:10

Re: irq/9-acpi interrupts and high cpu usage

#6 2025-12-15 20:57:15

Re: irq/9-acpi interrupts and high cpu usage

#7 2026-05-06 07:02:26

Re: irq/9-acpi interrupts and high cpu usage

#8 2026-05-06 10:23:46

Re: irq/9-acpi interrupts and high cpu usage

#9 2026-05-07 08:55:07

Re: irq/9-acpi interrupts and high cpu usage

#10 2026-05-07 16:13:03

Re: irq/9-acpi interrupts and high cpu usage

#11 2026-05-07 17:07:01

Re: irq/9-acpi interrupts and high cpu usage

#12 2026-05-11 18:23:44

Re: irq/9-acpi interrupts and high cpu usage

Board footer