another report of randomly freeze with unstable tsc

enihcam · 2021-08-18 04:36:31

latest report regarding to random freeze (at least 1 case per day) with unstable tsc:

[Aug17 12:47] rcu: INFO: rcu_preempt detected expedited stalls on CPUs/tasks: { 0-... 3-... } 4950563 jiffies s: 165 root: 0x9/.
[  +0.002957] rcu: blocking rcu_node structures:
[  +0.000041] Task dump for CPU 0:
[  +0.000001] task:kworker/0:0     state:R  running task     stack:    0 pid: 5265 ppid:     2 flags:0x00004008
[  +0.000004] Workqueue: events 0xffffffff815029e0
[  +0.000002] Call Trace:
[  +0.000003]  ? 0xffffffff815028c0
[  +0.000002]  ? 0xffffffff815029fc
[  +0.000001]  ? 0xffffffff810a8bf1
[  +0.000001]  ? 0xffffffff810a90f8
[  +0.000001]  ? 0xffffffff810a90b0
[  +0.000001]  ? 0xffffffff810ae14a
[  +0.000001]  ? 0xffffffff810ae020
[  +0.000001]  ? 0xffffffff8100106f
[  +0.000002] Task dump for CPU 3:
[  +0.000001] task:kworker/3:1     state:R  running task     stack:    0 pid:   38 ppid:     2 flags:0x00004008
[  +0.000003] Workqueue: events 0xffffffffa02424f0
[  +0.000001] Call Trace:
[  +0.000002]  ? 0xffffffffa0242591
[  +0.000001]  ? 0xffffffff810a8bf1
[  +0.000001]  ? 0xffffffff810a90f8
[  +0.000001]  ? 0xffffffff810a90b0
[  +0.000001]  ? 0xffffffff810ae14a
[  +0.000001]  ? 0xffffffff810ae020
[  +0.000001]  ? 0xffffffff8100106f
[Aug17 12:49] TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.
[  +0.000003] sched_clock: Marking unstable (49411118887439, 396353508)<-(49411596574654, -1621353)
[  -0.079652] clocksource: Checking clocksource tsc synchronization from CPU 1.
[  +0.000257] clocksource: Switched to clocksource hpet
[  +0.036592] br0: port 5(enp3s0) entered disabled state
[  +0.010741] igb 0000:03:00.0 enp3s0: Reset adapter

Kernel, RCU, clock and CPU context:

[  +0.000000] Linux version 5.10.59-xanmod1-1-lts (makepkg@archlinux) (gcc (GCC) 11.1.0, GNU ld (GNU Binutils) 2.36.1) #2 SMP PREEMPT Mon, 16 Aug 2021 00:30:08 +0000
[  +0.000000] Command line: BOOT_IMAGE=/BOOT/vmlinuz-linux-xanmod-lts root=/dev/sda rw rootflags=subvol=ROOT acpi_osi=Linux audit=0 init=/lib/systemd/systemd ipv6.disable=1 loglevel=3 mitigations=off quiet
...
[  +0.000063] rcu: Preemptible hierarchical RCU implementation.
[  +0.000001] rcu:      RCU priority boosting: priority 1 delay 0 ms.
[  +0.000002]   Trampoline variant of Tasks RCU enabled.
[  +0.000001]   Tracing variant of Tasks RCU enabled.
[  +0.000002] rcu: RCU calculated value of scheduler-enlistment delay is 100 jiffies.
[  +0.000024] NR_IRQS: 4352, nr_irqs: 1024, preallocated irqs: 16
[  +0.002684] Console: colour VGA+ 80x25
[  +0.000021] printk: console [tty0] enabled
[  +0.000006] ACPI: Core revision 20200925
[  +0.000429] clocksource: hpet: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 79635855245 ns
[  +0.000080] APIC: Switch to symmetric I/O mode setup
[  +0.004093] ..TIMER: vector=0x30 apic1=0 pin1=2 apic2=-1 pin2=-1
[  +0.004890] clocksource: tsc-early: mask: 0xffffffffffffffff max_cycles: 0x29ccd767b87, max_idle_ns: 440795223720 ns
[  +0.000006] Calibrating delay loop (skipped), value calculated using timer frequency.. 5799.77 BogoMIPS (lpj=2899886)
[  +0.000005] pid_max: default: 32768 minimum: 301
[  +0.000057] Mount-cache hash table entries: 8192 (order: 4, 65536 bytes, linear)
[  +0.000013] Mountpoint-cache hash table entries: 8192 (order: 4, 65536 bytes, linear)
[  +0.000323] mce: CPU0: Thermal monitoring enabled (TM1)
[  +0.000029] process: using mwait in idle threads
[  +0.000004] Last level iTLB entries: 4KB 64, 2MB 8, 4MB 8
[  +0.000001] Last level dTLB entries: 4KB 64, 2MB 0, 4MB 0, 1GB 4
[  +0.000006] Speculative Store Bypass: Vulnerable
[  +0.000006] SRBDS: Vulnerable
[  +0.000206] Freeing SMP alternatives memory: 24K
[  +0.000389] smpboot: CPU0: Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz (family: 0x6, model: 0x8e, stepping: 0x9)
[  +0.000120] Performance Events: PEBS fmt3+, Skylake events, 32-deep LBR, full-width counters, Intel PMU driver.
[  +0.000017] ... version:                4
[  +0.000001] ... bit width:              48
[  +0.000001] ... generic registers:      4
[  +0.000002] ... value mask:             0000ffffffffffff
[  +0.000001] ... max period:             00007fffffffffff
[  +0.000001] ... fixed-purpose events:   3
[  +0.000001] ... event mask:             000000070000000f
[  +0.000149] rcu: Hierarchical SRCU implementation.

I found another post talking about this https://bbs.archlinux.org/viewtopic.php?id=256576 but with no solutions.

switch to HPET does not resolve the issue. anyone any suggestions? thanks.

Last edited by enihcam (2021-08-18 04:38:01)

V1del · 2021-08-18 06:53:08

What happens if you remove the acpi_osi line? And make sure early microcode is set up: https://wiki.archlinux.org/title/Microcode

Last edited by V1del (2021-08-18 06:53:55)

enihcam · 2021-08-18 08:14:34

V1del wrote:

What happens if you remove the acpi_osi line? And make sure early microcode is set up: https://wiki.archlinux.org/title/Microcode

i'll try and see, but 'acpi_osi=Linux' has been existing since beginning (2 years ago) but freeze happened recently.

enihcam · 2021-08-19 01:30:07

enihcam wrote:

V1del wrote:
What happens if you remove the acpi_osi line? And make sure early microcode is set up: https://wiki.archlinux.org/title/Microcode
i'll try and see, but 'acpi_osi=Linux' has been existing since beginning (2 years ago) but freeze happened recently.

acpi_osi is removed, also the microcode is already early set up. system is still unstable, with same error message.

[Aug19 00:41] microcode: microcode updated early to revision 0xea, date = 2021-01-05
[  +0.000000] Linux version 5.10.59-xanmod1-1-lts (makepkg@archlinux) (gcc (GCC) 11.1.0, GNU ld (GNU Binutils) 2.36.1) #2 SMP PREEMPT Mon, 16 Aug 2021 00:30:08 +0000
[  +0.000000] Command line: BOOT_IMAGE=/BOOT/vmlinuz-linux-xanmod-lts root=/dev/sda rw rootflags=subvol=ROOT acpi_osi=Linux audit=0 init=/lib/systemd/systemd ipv6.disable=1 loglevel=3 mitigations=off quiet

V1del · 2021-08-19 02:09:20

Well at that point, I'd check for a BIOS/UEFI update. And maybe try a standard and non LTS kernel, can you reproduce there?

Arch Linux

#1 2021-08-18 04:36:31

another report of randomly freeze with unstable tsc

#2 2021-08-18 06:53:08

Re: another report of randomly freeze with unstable tsc

#3 2021-08-18 08:14:34

Re: another report of randomly freeze with unstable tsc

#4 2021-08-19 01:30:07

Re: another report of randomly freeze with unstable tsc

#5 2021-08-19 02:09:20

Re: another report of randomly freeze with unstable tsc

Board footer