> right now, I can outline what I did yesterday. If you
> want more detail, just ask and I'll provide precise
> data.
Yes, more details please.
SPARC or x86? What kind of chipset is used for the
USB host controller? The USB host controller has USB2
support, correct?
> Yesterday, I tried to copy data to an USB stick
> (documented two issues concerning this in the bugs
> mailing list). During this I did an "iostat -xzn 5"
> and got about 700k/s on an USB2 stick that easily can
> deliver multimegabyte transfer rates (on opensolaris
> bfu 20060320).
Did you connect the USB2 stick to an USB2 host controller port?
700k/s sounds as if you've connected it to an USB1 host
controller (and that would never get more than ~1000k/s anyway).
Check with "prtconf -D" and check that the "scsa2usb" device node
for the USB stick is listed as a child for an "ehci" (=USB2) controller.
You should find something like this:
...
pcie55,2928, instance #0 (driver name: ehci)
storage, instance #1 (driver name: scsa2usb)
disk, instance #6 (driver name: sd)
....
> The funny thing about this is that it wasn't pcfs related.
Is reading from the raw device slow, too?
I just tried this with an USB2 flash memory stick:
% time dd if=/vol/dev/rdsk/c3t0d0/flashdisk of=/dev/null bs=64k count=1000
1000+0 records in
1000+0 records out
0.00u 0.23s 0:04.26 5.3%
% bc
64*1024*1000/4.26
15384037
15.3 mb/sec. Looks OK.
This message posted from opensolaris.org
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
