"Unaligned handler failed, ret = -2" and other errors on C8000

"Unaligned handler failed, ret = -2" and other errors on C8000

[Rolf Eike Beer]

[-- Attachment #1.1: Type: text/plain, Size: 5054 bytes --]

I just found out my C8000 rebootet, this is the log:

[33718.498892] Unaligned handler failed, ret = -2
[33718.558778]       _______________________________
33718.558778]      < Your System ate a SPARC! Gah! >
33718.558778]       -------------------------------
33718.558778]              \   ^__^
33718.558778]                  (__)\       )\/\
33718.558778]                   U  ||----w |
33718.558778]                      ||     ||
[33718.950785] f951 (pid 921): Unaligned data reference (code 28)
[33719.026783] CPU: 2 PID: 921 Comm: f951 Not tainted 5.7.1-gentoo-parisc64 #1
[33719.118767] Hardware name: 9000/785/C8000
[33719.174768]
[33719.194765]      YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
[33719.254776] PSW: 00001000000000010000000000001000 Not tainted
[33719.330834] r00-03  0000000008010008 fffffff0f0e37000 fffffff0f0d75858 0000000000000000
[33719.438772] r04-07  fffffff0f0407580 fffffff0f0407180 fffffffffffffff9 0000000000000000
[33719.542780] r08-11  0000000000000001 8400000000800000 fffffff0f040c3c0 8400000000800000
[33719.646768] r12-15  0000000000000000 00000000f952860c 00000000f9528608 00000000f95283e0
[33719.754769] r16-19  00000000f95285a0 00000000f95283e4 00000000f95283b8 0000000000000000
[33719.858777] r20-23  0000000000003d60 00000000000001ea 0000000000000000 0000000000000000
[33719.966768] r24-27  fffffff0f0407a2c 0000000000000000 fffffff0f0e93508 fffffff0f0400000
[33720.070769] r28-31  fffffff0f0438e70 fffffff0f0407580 fffffff0f0407680 636e746c696e6974
[33720.174777] sr00-03  0000000006b76800 0000000000000000 0000000000000000 0000000006b76800
[33720.282768] sr04-07  0000000006b76800 0000000006b76800 0000000006b76800 0000000006b76800
[33720.390766]
[33720.410766] IASQ: 000000003ffffff0 000000003ffffff0 IAOQ: fffffff0f0d597d4 fffffff0f0d597d8
[33720.518776]  IIR: 52730240    ISR: 000000003ffff800  IOR: c00007f0f0407b4c
[33720.610769]  CPU:        2   CR30: 0000000168d1c000 CR31: ffffffffffffffff
[33720.702765]  ORIG_R28: 0000000000000000
[33720.750773]  IAOQ[0]: 0xfffffff0f0d597d4
[33720.802795]  IAOQ[1]: 0xfffffff0f0d597d8
[33720.854768]  RP(r2): 0xfffffff0f0d75858
[33720.906765] Backtrace:
[33720.938779]
[33720.958771] CPU: 2 PID: 921 Comm: f951 Not tainted 5.7.1-gentoo-parisc64 #1
[33720.962755] Hardware name: 9000/785/C8000
[33720.962755] Backtrace:
[33720.962755]  [<0000000040188c3c>] show_stack+0x5c/0x70
[33720.962755]  [<00000000405778e4>] dump_stack+0xe4/0x158
[33720.962755]  [<0000000040188e48>] die_if_kernel+0x1f0/0x3f0
[33720.962755]  [<000000004019ba04>] handle_unaligned+0x574/0xb30
[33720.962755]  [<0000000040189590>] handle_interruption+0x2f0/0xcb8
[33720.962755]  [<000000004018f080>] intr_check_sig+0x0/0x3c
[33720.962755]
[33721.594770] ---[ end trace 955e25bb36439323 ]---
[33777.274755] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[33777.274755]  (detected by 0, t=15002 jiffies, g=7042769, q=27)
[33777.274755] rcu: All QSes seen, last rcu_sched kthread activity 14999 (4303336614-4303321615), jiffies_till_next_fqs=1, root ->qsmask 0x0
[33777.274755] as              R  running task        0   918    908 0x00000014
[33777.274755] Backtrace:
[33777.274755]  [<0000000040188c3c>] show_stack+0x5c/0x70
[33777.274755]  [<00000000401ff5dc>] sched_show_task.part.0+0x1a4/0x1c8
[33777.274755]  [<00000000401f7918>] sched_show_task+0x48/0x50
[33777.274755]  [<00000000402415d8>] rcu_sched_clock_irq+0xd38/0xd68
[33777.274755]  [<0000000040248360>] update_process_times+0x80/0x110
[33777.274755]  [<000000004099da24>] timer_interrupt+0xb4/0x130
[33777.274755]  [<00000000402284dc>] __handle_irq_event_percpu+0xc4/0x270
[33777.274755]  [<00000000402286c0>] handle_irq_event_percpu+0x38/0xd8
[33777.274755]  [<0000000040230bf0>] handle_percpu_irq+0xb0/0xe8
[33777.274755]  [<0000000040227558>] generic_handle_irq+0x50/0x60
[33777.274755]  [<0000000040190340>] call_on_stack+0x18/0x24
[33777.274755]  [<000000004018a948>] execute_on_irq_stack+0x80/0x98
[33777.274755]  [<000000004018bb48>] do_cpu_irq_mask+0x2a0/0x340
[33777.274755]  [<000000004018f074>] intr_return+0x0/0xc
[33777.274755]
[33777.274755] rcu: rcu_sched kthread starved for 14999 jiffies! g7042769 f0x2 RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=3
[33777.274755] rcu: RCU grace-period kthread stack dump:
[33777.274755] rcu_sched       R  running task        0    10      2 0x00000000
[33777.274755] Backtrace:
[33777.274755]

There are some other oddities:

I upgraded from 5.4 using SLAB to 5.6/5.7 using SLUB a few weeks ago. This 
changed the permanent instability of the system (random crashes of processes) 
to a different level: the machine now locks up every other day without any 
relevant output. Config attached.

The missing [ at the beginning of the continuation lines also happen on other 
errors:

[28930.634792] do_page_fault() command='kwsysTestProces' type=15 address=0x00000004 in kwsysTestProcess[10000+c000]
28930.634792] trap #15: Data TLB miss fault
[28930.822838] CPU: 3 PID: 1442 Comm: kwsysTestProces Not tainted 5.7.1-gentoo-parisc64 #1

Is there a KERN_CONT missing somewhere?

Any ideas?

Eike

[-- Attachment #1.2: config.gz --]
[-- Type: application/gzip, Size: 15386 bytes --]

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 195 bytes --]