Date: Sat, 1 Jul 2006 17:07:16 +0200 From: Hans Petter Selasky <hselasky@c2i.net> To: thierry@herbelot.com Cc: freebsd-hackers@freebsd.org Subject: Re: High-speed transfers Message-ID: <200607011707.17055.hselasky@c2i.net> In-Reply-To: <200607011624.08249.thierry@herbelot.com> References: <200605271102.19799.hselasky@c2i.net> <200607011531.11061.hselasky@c2i.net> <200607011624.08249.thierry@herbelot.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday 01 July 2006 16:24, Thierry Herbelot wrote: > Le Saturday 1 July 2006 15:31, Hans Petter Selasky a =E9crit : > > Yes, but don't forget high-speed USB transfers. They require larger > > buffers. For example 1024 bytes for ULPT is too little. The interrupt > > rate will be so high, that it is unrealistic to transfer 20MB/s using > > 1024 byte interrupts. My rewritten ULPT now uses "2*(1<<17)" buffers. > > Hello, > > I wonder what kind of speed you are getting=20 It depends much on the equipment you are using. For example I have a Brothe= r=20 USB 2.0, Laser Printer, postscript compatible, and if I were to guess, it=20 eats something like 1MB/s. On the other hand I tested with an external USB= =20 2.0, 2.5" IDE adapter, and it transfers around 20MB/s. > : I would like to see improvements for reads (and writes) on=20 > standard endpoints, without having to resort to writing specific=20 > drivers (using ugen on the standard FreeBSD USB stack). > > One goal would be to achieve something like 25 to 30 Mbytes/s, sustained, > (finally getting to some interesting fraction of the peak USB2 data rate). I don't think that will be a problem, as long as you use large enough buffe= rs.=20 A buffer of 1/50 th of the transfer rate should suffice, from my experience. =2D-HPS
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200607011707.17055.hselasky>