low USB tranfer rate with external IDE

[Vlad Skvortsov]

M. Warner Losh wrote:
> In message: <46D47646.7080509 at 73rus.com>
>             Vlad Skvortsov <vss at 73rus.com> writes:
> : Andrey V. Elsukov wrote:
> : > Vlad Skvortsov wrote:
> : >
> : >> Any clues? Where should I look at? I'm concerned with 'USB revision 
> : >> 1.0' in the output -- could that be the problem?
> : >
> : > Do you have "device ehci" in your kernel config?
> : >
> : 
> : No, I don't. Should I? The man page says:
> : 
> : "The driver is not finished and is quite buggy."
>
> The man page takes a too pessimistic view of ehci.  It works for most
> people.
>
> Warner
>   

Ok, I have compiled a kernel with ehci in. Here is a snippet of dmesg 
output (right after rebooting the box):

da0 at umass-sim0 bus 0 target 0 lun 0
da0: <Maxtor 6 L250R0 BAJ4> Fixed Direct Access SCSI-0 device
da0: 40.000MB/s transfers
da0: 238475MB (488397168 512 byte sectors: 255H 63S/T 30401C)
umass0: BBB reset failed, TIMEOUT
umass0: BBB reset failed, TIMEOUT
umass0: Invalid CSW: sig 0x4628f8ff should be 0x53425355
(da0:umass-sim0:0:0:0): Synchronize cache failed, status == 0x4, scsi 
status == 0x0


Note that it displays high transfer rate this time. I've ran 'diskinfo 
-t' to check the actual rate; the seek tests have passed but transfers 
hung. 'dd' froze either:

13:31 [vss at bear] ~ > dd if=/dev/da0 of=/dev/null bs=1m count=10
^C

Nothing relevant in /var/log/messages. How can I debug it further?

-- 
Vlad Skvortsov, vss at 73rus.com, http://vss.73rus.com