Firewire/USB conflicting number of sectors for certain drives

[Barry Bouwsma]

Salut,

First a quick question -- I've noticed that a few soundcards exist
(external) with Firewire interfaces, and I wonder what the chances
of FreeBSD being able to support such a device might be, if anyone
might be able to comment.  Thanks...


Now, the issue at hand:

I have three different disks in front of me, all equipped with both
USB and Firewire connectors.  Two of the three report a different
number of sectors -- one fewer -- when connected via firewire, than
when connected via USB1 (UHCI and apparently OHCI though I know of
access problems with the latter; have not checked against EHCI).
The third disk, however, reports the same sectors in `dmesg', and
I initially set it up connected with Firewire, while the other drives
have been set up over USB1 as the hardware which I had at hand.

The two drives are Western Digital, hidden behind something as you
can see in the below dmesg snippets.  Here the reported sectors from
them:
USB:  da1: 238475MB (488397169 512 byte sectors: 255H 63S/T 30401C)
FW:   da1: 238475MB (488397168 512 byte sectors: 255H 63S/T 30401C)
and
USB:  da1: 190782MB (390721969 512 byte sectors: 255H 63S/T 24321C)
FW:   da1: 190782MB (390721968 512 byte sectors: 255H 63S/T 24321C)

Too bad I disklabel'ed and whatnot the disk when connected via USB,
so that any Firewire access fails to report the actual partition
where I expect to find the data (it now appears to extend beyond the
end of the disk).  Hmmm.
da1s1: slice extends beyond end of disk: truncating from 488397106 to 488397105
sectors

The third drive -- the first one I acquired -- is a Maxtor, and does
not have this issue.  I'm using kernel modules on FreeBSD 4.x, modules
built on 11.05.2004, which otherwise work splendidly, though I may have
one or two minor hacks compiled in.  I haven't tried more recent source
or anything based on -current.


It's probably a known issue that (as you see below), the drive attached
by USB is identified by what the actual drive is, whereas when connected
via firewire, it appears to be identified as the chipset that interfaces
from firewire to the drive's ATA -- at least with my modules.


Here's a bit of some `dmesg' concerning the drives, although it's
not comprehensive, in case there's something to be seen below:

exhibit 1 on usb:
umass0: ASSMANN Electronic GmbH product 0x3507, rev 2.00/0.01, addr 4
umass0:3:0:-1: Attached to scbus3
pass1 at umass-sim0 bus 0 target 0 lun 0
pass1: <WDC WD25 00JB-00GVA0 08.0> Fixed Direct Access SCSI-0 device
pass1: 1.000MB/s transfers
Creating DISK da1
da1 at umass-sim0 bus 0 target 0 lun 0
da1: <WDC WD25 00JB-00GVA0 08.0> Fixed Direct Access SCSI-0 device
da1: 1.000MB/s transfers
da1: 238475MB (488397169 512 byte sectors: 255H 63S/T 30401C)

exhibit 1 then connected via firewire:
pass1 at sbp0 bus 0 target 1 lun 0
pass1: <ASSMANN  0102> Fixed Simplified Direct Access SCSI-4 device
pass1: Serial Number \^_
pass1: 50.000MB/s transfers
Creating DISK da1
sbp0:1:0 sbp_cam_scan_lun
da1 at sbp0 bus 0 target 1 lun 0
da1: <ASSMANN  0102> Fixed Simplified Direct Access SCSI-4 device
da1: Serial Number \^_
da1: 50.000MB/s transfers
da1: 238475MB (488397168 512 byte sectors: 255H 63S/T 30401C)
da1s1: type 0xa5, start 63, end = 488397168, size 488397106 : OK
da1s1: slice extends beyond end of disk: truncating from 488397106 to 488397105
sectors
da1: raw partition size != slice size
da1: start 63, end 488397167, size 488397105
da1c: start 63, end 488397168, size 488397106


exhibit 2 connected via usb:
umass1: Generic USB 2.0 Storage Device, rev 2.00/0.01, addr 5
umass1:4:1:-1: Attached to scbus4
pass1 at umass-sim1 bus 1 target 0 lun 0
pass1: <WDC WD20 00JB-00GVA0 08.0> Fixed Direct Access SCSI-0 device
pass1: 1.000MB/s transfers
Creating DISK da1
da1 at umass-sim1 bus 1 target 0 lun 0
da1: <WDC WD20 00JB-00GVA0 08.0> Fixed Direct Access SCSI-0 device
da1: 1.000MB/s transfers
da1: 190782MB (390721969 512 byte sectors: 255H 63S/T 24321C)

and then via firewire:
da1 at sbp0 bus 0 target 1 lun 0
da1: < 1394 to ATA 2804> Fixed Simplified Direct Access SCSI-4 device
da1: Serial Number ^_
da1: 50.000MB/s transfers
da1: 190782MB (390721968 512 byte sectors: 255H 63S/T 24321C)


for reference, the Maxtor drive appears comparable with both:
usb:
umass1: Maxtor 5000XT v01.00.00, rev 2.00/1.00, addr 5
umass1: Get Max Lun not supported (STALLED)
umass1:4:1:-1: Attached to scbus4
pass1 at umass-sim1 bus 1 target 0 lun 0
pass1: <Maxtor 5000XT v01.00.00 0100> Fixed Direct Access SCSI-0 device
pass1: Serial Number A80A06AE
pass1: 1.000MB/s transfers
Creating DISK da1
da1 at umass-sim1 bus 1 target 0 lun 0
da1: <Maxtor 5000XT v01.00.00 0100> Fixed Direct Access SCSI-0 device
da1: Serial Number A80A06AE
da1: 1.000MB/s transfers
da1: 239371MB (490232832 512 byte sectors: 255H 63S/T 30515C)

firewire:
pass1 at sbp0 bus 0 target 0 lun 0
pass1: <Maxtor 5000XT  v1.00.00 0000> Fixed Simplified Direct Access SCSI-4 devi
ce
pass1: Serial Number A80A06AE
pass1: 50.000MB/s transfers
Creating DISK da1
sbp0:0:0 sbp_cam_scan_lun
da1 at sbp0 bus 0 target 0 lun 0
da1: <Maxtor 5000XT  v1.00.00 0000> Fixed Simplified Direct Access SCSI-4 device
da1: Serial Number A80A06AE
da1: 50.000MB/s transfers
da1: 239371MB (490232832 512 byte sectors: 255H 63S/T 30515C)


I'll be rebuilding my modules with the latest source Real Soon Now,
to see if this is still an issue, and would be happy to provide any
needed info.


thanks
barry bouwsma