Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 28 Jan 2007 19:02:01 -0500
From:      =?ISO-8859-1?Q?Sten_Daniel_S=F8rsdal?= <sten.daniel.sorsdal@gmail.com>
To:        Hans Petter Selasky <hselasky@c2i.net>
Cc:        freebsd-current@freebsd.org, Alexey Karagodov <karagodov@gmail.com>, Stefan Ehmann <shoesoft@gmx.net>
Subject:   Re: Interesting speed benchmarks
Message-ID:  <45BD3979.908@gmail.com>
In-Reply-To: <200701261536.48893.hselasky@c2i.net>
References:  <20070125.192448.-432840241.imp@bsdimp.com>	<cb5206420701260435s66e0687bnb467a42379d0a8d3@mail.gmail.com>	<200701261341.03742.shoesoft@gmx.net> <200701261536.48893.hselasky@c2i.net>

next in thread | previous in thread | raw e-mail | index | archive | help
Hans Petter Selasky wrote:
> On Friday 26 January 2007 13:41, Stefan Ehmann wrote:
>> On Friday 26 January 2007 13:35, Andrew Pantyukhin wrote:
>>> On 1/26/07, Stefan Ehmann <shoesoft@gmx.net> wrote:
>>>> On Friday 26 January 2007 11:00, Alexey Karagodov wrote:
>>>>> what manufacturer says about usb speeds?
>>>>> that is the question
>>>> Well, "up to 56MB/s" which is pretty much full USB2 speed.
>=20
> It is called high speed USB, and it can go up 53 MB/s with a payload of=
 512=20
> bytes per packet according to "Table 5-10. High-speed Bulk Transaction =

> Limits" in the USB 2.0 specification. The table does not say anything a=
bout=20
> whether this include bit-stuffing or not. If bit stuffing is not includ=
ed,=20
> then you have to divide this value by 1.20 approximately for the worst =
case,=20
> all 1's. 53 MB/s div 1.20 =3D 44 MB/s.
>=20
>>>> But writing it on the box doesn't mean the speed can actually be
>>>> reached.
>>>>
>>>> Benchmarking on windows might be interesting, but I don't know how t=
o
>>>> measure raw disk io on windows.
>>> Format the disk, copy a large file to/from it, divide
>>> its size by time spent, add the word "approximately" :-)
>> I'd rather not format a drive with my backups and other stuff on it :-=
)
>=20
> Results with the new USB stack*:
>=20
> Changing the interrupt delay from 2 microframes to 1 microframe gave me=
=20
> 2MBytes more per second on the EHCI controller.
>=20
> I connected two high speed "umass" capable devices to the same EHCI con=
troller=20
> on my computer, and did a "dd" on both devices at the same time, with a=
 block=20
> size of 131072 bytes.
>=20
> The one device transferred 22 MB/s. The other device transferred 16 MB/=
s.=20
> Summed up this yields 38 MB/s. Used alone these devices can transfer 27=
 MB/s=20
> and 20 MB/s. It seems clear that the EHCI controller is saturated at 38=
 MB/s.=20
>=20
> %dmesg |grep ehci
> ehci0: <Intel 82801DB/L/M (ICH4) USB 2.0 controller> mem 0xe0100000-0xe=
01003ff=20
> irq 10 at device 29.7 on pci0
> usb3: <Intel 82801DB/L/M (ICH4) USB 2.0 controller> on ehci0
> %
>=20

Just FYI.

I get about the same performance on my laptop running Windows XP when
moving between local 100 mb 5400 rpm ATA and 250 mb 7200 rpm USB2.0 disk.=

I have the same controller.

--=20
Sten Daniel S=F8rsdal




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?45BD3979.908>