On 2018-09-16 7:52 AM, Rolf Eike Beer wrote:
Kernel Fault: Code=6 (Instruction TLB miss fault) regs=(ptrval) (Addr=0014a00c)
CPU: 0 PID: 1 Comm: swapper Not tainted 4.18.7 #1
YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001000000000000000110 Not tainted
r00-03 00040006 107c9cf0 10148b90 4f034340
r04-07 10148000 dbd0d6d9 1085a000 00014f80
r08-11 00712000 10125010 00000004 10827020
r12-15 10826820 101453f0 00000010 f0400004
r16-19 4f0331b0 00800700 10832020 0000000f
r20-23 0082f0c0 fee003f8 fee003fd 00000000
r24-27 00000000 ffffffff 00000000 107b7020
r28-31 00000001 00000037 4f034380 00000060
sr00-03 00000000 00000000 00000000 00000000
sr04-07 00000000 00000000 00000000 00000000
IASQ: 00000000 00000000 IAOQ: 0014a00c 0014a010
IIR: 43ffff80 ISR: 4f0344c0 IOR: 00000000
CPU: 0 CR30: 4f034000 CR31: ffffdffe
ORIG_R28: 4f0344c0
IAOQ[0]: 0x14a00c
IAOQ[1]: 0x14a010
RP(r2): flush_tlb_all+0x2c/0xac
Backtrace:
[<10103b50>] 0x10103b50
[<10103ff8>] 0x10103ff8
[<101481d4>] do_one_initcall+0x58/0x1a4
[<10101078>] 0x10101078
[<1063958c>] kernel_init+0x1c/0x140
[<1014f01c>] end_fault_vector+0x1c/0x24
Not much has changed in this code. Is this fault consistent from a cold
start? I've seen
some cases where problems occur in a reboot.
I'm guessing this is a 32-bit UP build?
You could disassemble the routine around this address 0x14a00c to see
what instruction
(4f0344c0) triggered the fault. You also could find the function that
flush_tlb_all
thought it was calling.
Dave
--
John David Anglin dave.anglin@xxxxxxxx
