> Though SERIRQ is a really good hint. Let me look into the details of that a
> bit more.
Maybe I have more hints to you:
- nolapic=1 also helps
- preventing i915.ko from being loaded also helps
I checked this on 4.18.17.
Why did we came at the i915? After running stress-ng for some seconds
(or any program that opens/closes /dev/ttyS*) I got this:
root@test:~# dmesg | grep do_IRQ
[ 15.627047] do_IRQ: 1.39 No irq handler for vector
[ 15.747112] do_IRQ: 1.39 No irq handler for vector
[ 15.747228] do_IRQ: 1.39 No irq handler for vector
[ 15.867514] do_IRQ: 1.39 No irq handler for vector
[ 15.926858] do_IRQ: 1.39 No irq handler for vector
[ 15.926987] do_IRQ: 1.39 No irq handler for vector
[ 16.047095] do_IRQ: 1.39 No irq handler for vector
[ 16.047209] do_IRQ: 1.39 No irq handler for vector
[ 16.167521] do_IRQ: 1.39 No irq handler for vector
Then I looked at this vector. I made sure the serial ports are both open
with a cat+redirection.
root@test:/sys/kernel/debug/irq/irqs# grep 39 *
3: Vector: 39
I looked at this file and with my limited knowledge I couldn't find
anything wrong there:
root@test:/sys/kernel/debug/irq/irqs# cat 3
handler: handle_edge_irq
device: (null)
status: 0x00000000
istate: 0x00000000
ddepth: 0
wdepth: 0
dstate: 0x05400200
IRQD_ACTIVATED
IRQD_IRQ_STARTED
IRQD_SINGLE_TARGET
IRQD_CAN_RESERVE
node: 0
affinity: 0-1
effectiv: 1
pending:
domain: IO-APIC-0
hwirq: 0x3
chip: IO-APIC
flags: 0x10
IRQCHIP_SKIP_SET_WAKE
parent:
domain: VECTOR
hwirq: 0x3
chip: APIC
flags: 0x0
Vector: 39
Target: 1
move_in_progress: 0
is_managed: 0
can_reserve: 1
has_reserved: 0
cleanup_pending: 0
Then I thought "Maybe there's an off-by-one error somewhere". I checked
at other vectors. No one used vector 40. But maybe vector 38?
root@test:/sys/kernel/debug/irq/irqs# grep 38 *
44: Vector: 38
45: Vector: 38
The output of those two files is:
root@test:/sys/kernel/debug/irq/irqs# cat 44
handler: handle_edge_irq
device: 0000:00:1b.0
status: 0x00000000
istate: 0x00000000
ddepth: 0
wdepth: 0
dstate: 0x01400200
IRQD_ACTIVATED
IRQD_IRQ_STARTED
IRQD_SINGLE_TARGET
node: -1
affinity: 0-1
effectiv: 1
pending:
domain: PCI-MSI-2
hwirq: 0x6c000
chip: PCI-MSI
flags: 0x10
IRQCHIP_SKIP_SET_WAKE
parent:
domain: VECTOR
hwirq: 0x2c
chip: APIC
flags: 0x0
Vector: 38
Target: 1
move_in_progress: 0
is_managed: 0
can_reserve: 0
has_reserved: 0
cleanup_pending: 0
root@test:/sys/kernel/debug/irq/irqs# cat 45
handler: handle_edge_irq
device: 0000:00:02.0
status: 0x00000000
istate: 0x00000000
ddepth: 0
wdepth: 0
dstate: 0x01400200
IRQD_ACTIVATED
IRQD_IRQ_STARTED
IRQD_SINGLE_TARGET
node: -1
affinity: 0-1
effectiv: 0
pending:
domain: PCI-MSI-2
hwirq: 0x8000
chip: PCI-MSI
flags: 0x10
IRQCHIP_SKIP_SET_WAKE
parent:
domain: VECTOR
hwirq: 0x2d
chip: APIC
flags: 0x0
Vector: 38
Target: 0
move_in_progress: 0
is_managed: 0
can_reserve: 0
has_reserved: 0
cleanup_pending: 0
According to lspci ...
root@test:/sys/kernel/debug/irq/irqs# lspci -nn
00:00.0 Host bridge [0600]: Intel Corporation Haswell-ULT DRAM Controller [8086:0a04] (rev 0b)
00:02.0 VGA compatible controller [0300]: Intel Corporation Haswell-ULT Integrated Graphics Controller [8086:0a06] (rev 0b)
00:14.0 USB controller [0c03]: Intel Corporation 8 Series USB xHCI HC [8086:9c31] (rev 04)
00:16.0 Communication controller [0780]: Intel Corporation 8 Series HECI #0 [8086:9c3a] (rev 04)
00:19.0 Ethernet controller [0200]: Intel Corporation Ethernet Connection I218-LM [8086:155a] (rev 04)
00:1b.0 Audio device [0403]: Intel Corporation 8 Series HD Audio Controller [8086:9c20] (rev 04)
00:1c.0 PCI bridge [0604]: Intel Corporation 8 Series PCI Express Root Port 1 [8086:9c10] (rev e4)
00:1d.0 USB controller [0c03]: Intel Corporation 8 Series USB EHCI #1 [8086:9c26] (rev 04)
00:1f.0 ISA bridge [0601]: Intel Corporation 8 Series LPC Controller [8086:9c43] (rev 04)
00:1f.2 SATA controller [0106]: Intel Corporation 8 Series SATA Controller 1 [AHCI mode] [8086:9c03] (rev 04)
00:1f.3 SMBus [0c05]: Intel Corporation 8 Series SMBus Controller [8086:9c22] (rev 04)
vector 44 is audio and vector 45 is i915. I started with removing i915
first. After a reboot the behavior of the system changed: the "no vector
for"-message was gone.
I added i915 again, rebooted, and the error was back.
root@test:/sys/kernel/debug/irq/irqs# cat /proc/interrupts
CPU0 CPU1
0: 8 0 IO-APIC 2-edge timer
1: 0 5 IO-APIC 1-edge i8042
3: 0 39 IO-APIC 3-edge ttyS1
4: 0 39 IO-APIC 4-edge
8: 1 0 IO-APIC 8-edge rtc0
9: 0 4 IO-APIC 9-fasteoi acpi
12: 7 0 IO-APIC 12-edge i8042
18: 0 0 IO-APIC 18-fasteoi i801_smbus
23: 33 0 IO-APIC 23-fasteoi ehci_hcd:usb1
40: 0 0 PCI-MSI 458752-edge PCIe PME, pciehp
41: 0 3618 PCI-MSI 512000-edge ahci[0000:00:1f.2]
42: 0 140 PCI-MSI 327680-edge xhci_hcd
43: 1711 0 PCI-MSI 409600-edge eth0
44: 0 343 PCI-MSI 442368-edge snd_hda_intel:card0
45: 119 0 PCI-MSI 32768-edge i915
NMI: 0 0 Non-maskable interrupts
LOC: 4467 5445 Local timer interrupts
SPU: 0 0 Spurious interrupts
PMI: 0 0 Performance monitoring interrupts
IWI: 1 1 IRQ work interrupts
RTR: 0 0 APIC ICR read retries
RES: 620 511 Rescheduling interrupts
CAL: 1189 308 Function call interrupts
TLB: 13 8 TLB shootdowns
TRM: 0 0 Thermal event interrupts
THR: 0 0 Threshold APIC interrupts
MCE: 0 0 Machine check exceptions
MCP: 2 3 Machine check polls
ERR: 0
MIS: 0
PIN: 0 0 Posted-interrupt notification event
NPI: 0 0 Nested posted-interrupt event
PIW: 0 0 Posted-interrupt wakeup event
PS: I'm okay with testing this on 4.20-rc1 or any git tree you throw at me...
