[Discuss] USB speeds

[gaf at blu.org (Jerry Feldman)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Another benchmark. This is from a 4GB image to a microSD on my home system:
At home I tested using the command 'dd if=ph-hc-dualboot-sd.img
of=/dev/sdd bs=1M
Copied the results from time, but I can't find it in my email.
Essentially it took 1 hour clock time.
Doing the exact same thing on Windows XP on my laptop took under 30
minutes (it was completed when I looked since there is no time(1) I am
aware of on Windows. I booted the Trisquel Live CD, mounted the Windows
drive and ran 'time dd if=ph-hc-dualboot-sd.img of=/dev/sdb bs=1M'
This ran in 14 minutes elapsed time.
The bottom line is that on my home system it appears to be running at
USB1 speeds.
This is the LSPCI for my Lenovo T60
00:1d.0 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI
Controller #1 (rev 02) (prog-if 00 [UHCI])
00:1d.1 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI
Controller #2 (rev 02) (prog-if 00 [UHCI])
00:1d.2 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI
Controller #3 (rev 02) (prog-if 00 [UHCI])
00:1d.3 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI
Controller #4 (rev 02) (prog-if 00 [UHCI])
00:1d.7 USB Controller: Intel Corporation N10/ICH 7 Family USB2 EHCI
Controller (rev 02) (prog-if 20 [EHCI])

And the LSPCI for my Penguin system:
00:02.0 USB Controller: nVidia Corporation MCP55 USB Controller (rev a1)
(prog-if 10 [OHCI])
00:02.1 USB Controller: nVidia Corporation MCP55 USB Controller (rev a2)
(prog-if 20 [EHCI])

However, it was my impression that USB2 was EHCI.


On 08/29/2011 03:39 PM, Jerry Feldman wrote:
>
> Just ran a test on my system with no BS specified (eg 512) and BS=1MB
> 4:22.87elapsed using the default, 1:49.50elapsed using BS=1MB
>
>
>
>
> On 08/29/2011 03:12 PM, Jerry Feldman wrote:
>
> > On 08/29/2011 03:05 PM, Rich Pieri wrote:
> >> Two factors come immediately to mind, neither of which have anything to
> > do with the bus.
>
> >> On Unix, file systems are mounted asynchronously by default. I've seen
> > this cause horrible performance degradation on USB flash media. Try
> > mounting sync instead.
>
> >> You neglected to specify a block size so dd defaults to 512 byte
> > blocks. This is horribly inefficient. Try powers of 2 starting at 1MB to
> > find an optimal value.
>
> > When I use dd to copy an image I make sure that the device is not
> > mounted. Note that I am copying to the device.
> > I think that when I copied the image, I did use bs=1MB. As I mentioned,
> > I will plan to run a benchmark test tomorrow in my office if all my
> > servers come up ok.
>
>
> > _______________________________________________
> > Discuss mailing list
> > Discuss at blu.org
> > http://lists.blu.org/mailman/listinfo/discuss
>
>
>

- -- 
Jerry Feldman <gaf at blu.org>
Boston Linux and Unix
PGP key id:3BC1EB90
PGP Key fingerprint: 49E2 C52A FC5A A31F 8D66  C0AF 7CEA 30FC 3BC1 EB90
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEVAwUBTl6PnXzqMPw7weuQAQJStAf/f+l/468dYix387OZC/Rs9Vw7wr4sSMEz
BT7JyHDv6jZNw+41jBI5Wt2W+US5ZYBrCfnAfvtDPZ+8qwXBVBTA3RXiPEUpmeIZ
GyiOhevcIMvgxC2dLkc0W5JccJtJCO0J43LgwhEnqSyDTqFNH5qdzF3dGJQ2bpL3
wkGdESR3ignc8nk4jsk28x4h9Xzwe7ADmmQKdSbnLLi0Jt2UTZOEnLDfij4NTD+q
r91aNd56Sb5M9tNA614FVs37B9KSigshIGgDw5/R5qzJeZMjAB9I2rg4ntrCd6br
XQ8Mr1gPxu5BYCiDIRSIMbkBtV7zWdonkQlSXtFVKm2d3FzTQlPpgQ==
=Y4Lp
-----END PGP SIGNATURE-----