Help making sense of firewire disk enclosure dilema
dwhite at gumbysoft.com
Mon Aug 23 10:12:42 PDT 2004
On Sun, 22 Aug 2004, George Hartzell wrote:
> The enclosure is smeared across the kitchen table at the moment (see
> below), so I can't give it's exact reaction to suggestion, but I have
> some representative output.
> In the course of mucking around, I've tried various combinations of
> 'fwcontrol -r' and 'camcontrol rescan all'. I'm not sure which caused
> which part of the dmesg output below, but it might be interesting:
> sbp0:0:0 request timeout(cmd orb:0x163ca634) ... agent reset
> sbp0:0:0 request timeout(cmd orb:0x163ca76c) ... target reset
> fwohci0: BUS reset
> fwohci0: node_id=0xc800ffc0, gen=6, CYCLEMASTER mode
> firewire0: 1 nodes, maxhop <= 0, cable IRM = 0 (me)
> firewire0: bus manager 0 (me)
> sbp0:0:0 request timeout(cmd orb:0x163ca9dc) ... reset start
> fwohci0: txd err= 3 miss Ack err
> sbp0:0:0 sbp_reset_start failed: resp=22
> firewire0: split transaction timeout dst=0xffc0 tl=0x30 state=4
> sbp0:0:0 sbp_reset_start failed: resp=60
I've seen this before, if one of the nodes locks up during negotiation. My
Athlon box with APIC enabled has a tendnecy to do that when the machine it
was cabled to for debugging would reboot.
> I figured out how to ask fwcontrol to tell me more about the device
> inside the enclosure, and discovered that it's a Prolific PL-3507. A
> little googling about suggests that it's a well known PITA device, e.g.
> So, I'm left to decide whether to muck with the firmware or just chuck
I'd give up now. :)
> Are there "quirks" for firewire devices like there used to be (are?)
> for usb devices?
Not that I'm aware of, but I don't know if firewire supports device/vendor
IDs like USB that can be used to match the quirk.
Doug White | FreeBSD: The Power to Serve
dwhite at gumbysoft.com | www.FreeBSD.org
More information about the freebsd-current