Some necessity running scsi_target that I have observed ( 10+ times )

[Jacky Xie]

1. data transfer: total about 300GBytes - 1000GBytes) before scsi_target 
crash.


2. machine running scsi_target crashed in targdone():

Fatal trap 12: page fault while in kernel mode
fatal virtual address = 0x2a
fatal code = supervisor write, page not present
instructioin pointer = 0x8:0xe12c3a8
stack pointer = 0x10: 0xd49a7d48
frame pointer = 0x10: 0xd49a7d50
code segment = base 0x0, limit 0xfffff, type 0x16
                        DPL 0, pres 1, def32 1, gran 1
processor flags = trace trap, interrupt enabled, kernel IOPL=0
current process = 26962(scsi_target)
interrupt mask =
kernel: type 12 trap, code = 0
Breakpoint at targdone + 0x84: movl %eax, 0x20 (%edx)

3. machine running "dd" to read/write "FreeBSD Emulated Disk" -- /dev/da0 
report kernel warnings:

Sep 26 12:25:37 IA /kernel: (probe0:ahc0:0:5:0): SCB 0xe - timed out
Sep 26 12:25:37 IA /kernel: ahc0: Dumping Card State in Command phase, at
SEQADDR 0x168
Sep 26 12:25:37 IA /kernel: ACCUM = 0x80, SINDEX = 0xa0, DINDEX = 0xe4, AR
G_2 = 0x3
Sep 26 12:25:37 IA /kernel: HCNT = 0x0 SCBPTR = 0x0
Sep 26 12:25:37 IA /kernel: SCSISEQ = 0x12, SBLKCTL = 0xa
Sep 26 12:25:37 IA /kernel: DFCNTRL = 0x4, DFSTATUS = 0x89
Sep 26 12:25:37 IA /kernel: LASTPHASE = 0x80, SCSISIGI = 0x84, SXFRCTL0 =
0x88
Sep 26 12:25:37 IA /kernel: SSTAT0 = 0x7, SSTAT1 = 0x0
Sep 26 12:25:37 IA /kernel: SCSIPHASE = 0x0
Sep 26 12:25:37 IA /kernel: STACK == 0x175, 0x160, 0xe7, 0x34
Sep 26 12:25:37 IA /kernel: SCB count = 20
Sep 26 12:25:37 IA /kernel: Kernel NEXTQSCB = 15
Sep 26 12:25:37 IA /kernel: Card NEXTQSCB = 15
Sep 26 12:25:37 IA /kernel: QINFIFO entries:
Sep 26 12:25:37 IA /kernel: Waiting Queue entries:
Sep 26 12:25:37 IA /kernel: Disconnected Queue entries:
Sep 26 12:25:37 IA /kernel: QOUTFIFO entries:
Sep 26 12:25:37 IA /kernel: Sequencer Free SCB List: 1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Sep 26 12:25:37 IA /kernel: Sequencer SCB Info: 0(c 0x0, s 0x58, l 0, t 0x
e) 1(c 0x0, s 0xff, l 255, t 0xff) 2(c 0x0, s 0xff, l 255, t 0xff) 3(c 0x0
, s 0xff, l 255, t 0xff) 4(c 0x0, s 0xff, l 255, t 0xff) 5(c 0x0, s 0xff,
l 255, t 0xff) 6(c 0x0, s 0xff, l 255, t 0xff) 7(c 0x0, s 0xff, l 255, t 0
xff) 8(c 0x0, s 0xff, l 255, t 0xff) 9(c 0x0, s 0xff, l 255, t 0xff) 10(c
0x0, s 0xff, l 255, t 0xff) 11(c 0x0, s 0xff, l 255, t 0xff) 12(c 0x0, s 0
xff, l 255, t 0xff) 13(c 0x0, s 0xff, l 255, t 0xff) 14(c 0x0, s 0xff, l 2
55, t 0xff) 15(c 0x0, s 0xff, l 255, t 0xff) 16(c 0x0, s 0xff, l 255, t 0x
ff) 17(c 0x0, s 0xff, l 255, t 0xff) 18(c 0x0, s 0xff, l 255, t 0xff) 19(c
0x0, s 0xff, l 255, t 0xff) 20(c 0x0, s 0xff, l 255, t 0xff) 21(c 0x0, s
0xff, l 255, t 0xff) 22(c 0x0, s 0xff, l 255, t 0xff) 23(c 0x0, s 0xff, l
255, t 0xff) 24(c 0x0, s 0xff, l 255, t 0xff) 25(c 0x0, s 0xff, l 255, t 0
xff) 26(c 0x0, s 0xff, l 255, t 0xff) 27(c 0x0, s 0xff, l 255, t 0xff) 28(
c 0x0, s 0xff, l 255, t 0xff) 29(c 0x0, s 0xff, l 255, t 0xff) 30(c 0x0, s
0xff,
Sep 26 12:25:37 IA /kernel: 0xff) 31(c 0x0, s 0xff, l 255, t 0xff)
Sep 26 12:25:37 IA /kernel: Pending list: 14(c 0x0, s 0x58, l 0)
Sep 26 12:25:37 IA /kernel: Kernel Free SCB list: 16 17 18 19 0 1 2 3 4 5
6 7 8 9 13 12 11 10
Sep 26 12:25:37 IA /kernel: Untagged Q(5): 14
Sep 26 12:25:37 IA /kernel: sg[0] - Addr 0xf1c0 : Length 32
Sep 26 12:25:37 IA /kernel: (probe0:ahc0:0:5:0): BDR message in message bu
ffer

Sep 26 12:25:39 IA /kernel: (probe0:ahc0:0:5:0): SCB 0xe - timed out
Sep 26 12:25:39 IA /kernel: ahc0: Dumping Card State in Command phase, at
SEQADDR 0x168
......
SCB 0xf - timed out
ahc0: Dumping Card state while idle
......
SCB 0xe - timed out
ahc0: Dumping Card state while idle
......
SCB 0xf - timed out
ahc0: Dumping Card state while idle
......

I think some little BUG exists in Kernel scsi_target or Userland 
scsi_target.
Any hints?
Thanks.

---------Jacky---------
xjrcool at hotmail.com
---------^-^-----------

_________________________________________________________________