Raspberry Pi: still getting prefetch aborts
George Mitchell
george+freebsd at m5p.com
Thu Dec 26 00:09:32 UTC 2013
FreeBSD 10.0-PRERELEASE (RPI-B) #0 r259866M: Wed Dec 25 17:26:28 EST 2013
(M because I selected serial output in RPI-B)
/usr/ports is NFS mounted from elsewhere.
On the theory that perhaps my hardware is flaky, I bought another RPi
and got the same results on it. This sequence is now close to 100%
reproducible. Surely this is happening to someone else! And if not,
please point me to your working image. -- George
root at pi:~ # cd /usr/ports/ports-mgmt/pkg
root at pi:/usr/ports/ports-mgmt/pkg # make
Fatal kernel mode prefetch abort at 0x4278f502
trapframe: 0xdc369f98, spsr=60000033
r0 =00000000, r1 =000000c0, r2 =600000d3, r3 =60000013
r4 =c22efc00, r5 =00000000, r6 =c22e2140, r7 =c0675858
r8 =c25bb800, r9 =c2635100, r10=c25bb8e4, r11=0004240c
r12=00000000, ssp=dc369fe8, slr=c046405c, pc =4278f502
[ thread pid 876 tid 100067 ]
Stopped at 0x4278f502: address 0x4278f502 is invalid
*** error reading from address 4278f502 ***
KDB: reentering
KDB: stack backtrace:
db_trace_self() at db_trace_self
pc = 0xc04607e8 lr = 0xc012ebfc (db_trace_self_wrapper+0x30)
sp = 0xdc369b40 fp = 0xdc369c58
r10 = 0xc0547ac0
db_trace_self_wrapper() at db_trace_self_wrapper+0x30
pc = 0xc012ebfc lr = 0xc026a238 (kdb_reenter+0x60)
sp = 0xdc369c60 fp = 0xdc369c68
r4 = 0xc05884e0 r5 = 0xc05884c4
r6 = 0xdc369c70 r7 = 0x4278f502
kdb_reenter() at kdb_reenter+0x60
pc = 0xc026a238 lr = 0xc012b4f8 (db_get_value+0x40)
sp = 0xdc369c70 fp = 0xdc369c88
r4 = 0x00000004 r5 = 0x00000000
db_get_value() at db_get_value+0x40
pc = 0xc012b4f8 lr = 0xc045f25c (db_disasm_read_word+0x14)
sp = 0xdc369c90 fp = 0xdc369c90
r4 = 0x4278f502 r5 = 0xc0488f00
r6 = 0xc05884f0 r7 = 0xdc369f98
db_disasm_read_word() at db_disasm_read_word+0x14
pc = 0xc045f25c lr = 0xc0460c00 (disasm+0x20)
sp = 0xdc369c98 fp = 0xdc369d80
disasm() at disasm+0x20
pc = 0xc0460c00 lr = 0xc045f240 (db_disasm+0x18)
sp = 0xdc369d88 fp = 0xdc369d88
r4 = 0x4278f502 r5 = 0xdc369da8
r6 = 0xc05884f0 r7 = 0xdc369f98
r8 = 0xdc369f98 r9 = 0xc05884c4
r10 = 0xc0547ac0
db_disasm() at db_disasm+0x18
pc = 0xc045f240 lr = 0xc012d5a8 (db_print_loc_and_inst+0x2c)
sp = 0xdc369d90 fp = 0xdc369d98
db_print_loc_and_inst() at db_print_loc_and_inst+0x2c
pc = 0xc012d5a8 lr = 0xc012ed18 (db_trap+0xc0)
sp = 0xdc369da0 fp = 0xdc369ec0
r4 = 0x00000000
db_trap() at db_trap+0xc0
pc = 0xc012ed18 lr = 0xc026a4ec (kdb_trap+0xd4)
sp = 0xdc369ec8 fp = 0xdc369ee8
r4 = 0x00000000 r5 = 0x00000000
r6 = 0xc05884f0 r7 = 0xdc369f98
kdb_trap() at kdb_trap+0xd4
pc = 0xc026a4ec lr = 0xc0471ef0 (dab_fatal+0x174)
sp = 0xdc369ef0 fp = 0xdc369f08
r4 = 0xdc369f98 r5 = 0x200001d3
r6 = 0x4278f502 r7 = 0x00000000
r8 = 0xdc369f98 r9 = 0xc2635100
r10 = 0xc26fd000
dab_fatal() at dab_fatal+0x174
pc = 0xc0471ef0 lr = 0xc04721ac (prefetch_abort_handler+0x254)
sp = 0xdc369f10 fp = 0xdc369f90
r4 = 0x4278f502 r5 = 0xc272c320
r6 = 0xc22e2140 r7 = 0xc0675858
prefetch_abort_handler() at prefetch_abort_handler+0x254
pc = 0xc04721ac lr = 0xc04623e0 (exception_exit)
sp = 0xdc369f98 fp = 0x0004240c
r4 = 0xc22efc00 r5 = 0x00000000
r6 = 0xc22e2140 r7 = 0xc0675858
r8 = 0xc25bb800 r9 = 0xc2635100
r10 = 0xc25bb8e4
exception_exit() at exception_exit
pc = 0xc04623e0 lr = 0xc046405c (spinlock_exit+0x18)
sp = 0xdc369fec fp = 0x0004240c
r0 = 0x00000000 r1 = 0x000000c0
r2 = 0x600000d3 r3 = 0x60000013
r4 = 0xc22efc00 r5 = 0x00000000
r6 = 0xc22e2140 r7 = 0xc0675858
r8 = 0xc25bb800 r9 = 0xc2635100
r10 = 0xc25bb8e4 r12 = 0x00000000
Unable to unwind into user mode
db> show proc 876
Process 876 (make) at 0xc26fd000:
state: NORMAL
uid: 0 gids: 0, 5
parent: pid 873 at 0xc261a640
ABI: FreeBSD ELF32
arguments: make
threads: 1
100067 Run CPU 0 make
db> show thread 100067
Thread 100067 at 0xc272c320:
proc (pid 876): 0xc26fd000
name: make
stack: 0xdc369000-0xdc36afff
flags: 0x1002004 pflags: 0
state: RUNNING (CPU 0)
priority: 140
container lock: sched lock (0xc05935c0)
db>
More information about the freebsd-arm
mailing list