firewire disk stopped working

[Daniel Thiele]

Boris Kotzev wrote:
>> Well, I certainly have made a mess of this.  I have been fairly sick the
>> last couple of days could only come up with two things for you to try.
>>
>> 1.  set "sysctl hw.firewire.hold_count=3" and see if that resolves your
>> issue.
>>
>> 2.  Only set "debug.firewire_debug=2" ... this will be a very noisy
>> debug level.  Just capture what you can and email it back.
>>
>> Sean
> 
> Hi,
> 
> Setting "sysctl hw.firewire.hold_count=3" does not help - the disk still does not show up.
> 
> I am attaching to this email the dmesg output I captured after setting
> "debug.firewire_debug=2"  (debug.sbp_debug was set to 0).
> 

Hi,

I am having the same problem as Boris. My external firewire disk is not
recognized anymore.

I updated my machine from 7.1-STABLE to 8.0-CURRENT around Feb 24th.
With 7.1-STABLE the external disk was working fine. With 8.0-CURRENT,
however, it does not even find its way to /dev as daX and I get the
following errors after attaching the disk:

Obj: Data buffer, Error: Type error
sbp0: sbp_scsi_status:sbp0:0:0 unknown scsi status 0x0
sbp0: sbp0:0:0


I applied the last patch Sean provided against the firwire.c from Feb
24th and like in Boris' case this did not solve the problem.

A normal dmesg with debug.sbp_debug = 1 and debug.firewire_debug = 1
with the unpatched firwire.c can be found here:
http://www-public.tu-bs.de:8080/~y0023183/FreeBSD/unpatched_dmesg.txt

A verbose dmesg with hw.firewire.hold_count = 3 and
debug.firewire_debug = 2 is located here:
http://www-public.tu-bs.de:8080/~y0023183/FreeBSD/patched_dmesg-v_hold_count-3_firewire_debug-2.txt

If you need more verbose or additional logs just drop me a line and I
will try to provide them.

Best regards,

Daniel


P.S.: Sorry for messing up this threads hierarchy by replying to Boris' 
email and not to Seans, but I subscribed to freebsd-firewire@ on Feb 
25th, thus Boris' email was the only one I had so far to reply to.


Additional information:

uname -a:
FreeBSD impala.vnws.lan 8.0-CURRENT FreeBSD 8.0-CURRENT #0: Wed Feb 25 
17:13:30 CET 2009 
dthiele at impala.vnws.lan:/usr/obj/usr/src/sys/kernel_8Xv0  i386

I applied Bruno's patches against this CURRENT's firewire.c.


The interesting part from pciconf -lv:

fwohci0 at pci0:21:0:1:    class=0x0c0010 card=0x20c717aa chip=0x08321180 
rev=0x04 hdr=0x00
     vendor     = 'Ricoh Company, Ltd.'
     device     = 'unknown IEEE 1394 (4 pin firewire) chip)'
     class      = serial bus
     subclass   = FireWire


And fwcontrol -p:

=== base register ===
0x03 0x05 0xe1 0x41 0xc8 0x03 0x00 0x00
Physical_ID:0  R:1  CPS:1
RHB:0  IBR:0  Gap_Count:5
Extended:7 Num_Ports:1
PHY_Speed:2 Delay:1
LCtrl:1 C:1 Jitter:1 Pwr_Class:0
WDIE:0 ISBR:0 CTOI:0 CPSI:0 STOI:0 PEI:0 EAA:1 EMC:1
Max_Legacy_SPD:0 BLINK:0 Bridge:0
Page_Select:0 Port_Select0

=== page 0 port 0 ===
0xf8 0x00 0x00 0x00 0x00 0x00 0x00 0x00
Astat:3 BStat:3 Ch:1 Con:0 RXOK:0 Dis:0
Negotiated_speed:0 PIE:0 Fault:0 Stanby_fault:0 Disscrm:0 B_Only:0
DC_connected:0 Max_port_speed:0 LPP:0 Cable_speed:0
Connection_unreliable:0 Beta_mode:0
Port_error:0x0
Loop_disable:0 In_standby:0 Hard_disable:0

=== page 1 ===
0x01 0x00 0x00 0x00 0x74 0x00 0x83 0x20
Compliance:1
Vendor_ID:0x000074
Product_ID:0x008320