Well, probably not today, but I hope that it
can soon.

Bruce?

~Randy
___________________________________________________
|Randy Dunlap     Intel Corp., DAL    Sr. SW Engr.|
|randy.dunlap.at.intel.com            503-696-2055|
|NOTE:  Any views presented here are mine alone   |
|and may not represent the views of my employer.  |
|_________________________________________________|

> -----Original Message-----
> From: Pavel Machek [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, March 29, 2000 2:49 AM
> To: Dunlap, Randy; '[EMAIL PROTECTED]'
> Subject: Re: [linux-usb] FW: usbdevfs comparison (Rio500)
> 
> > Hi everyone.  Just performed a comparison by dumping Sting's Mercury
> > Falling CD to the Rio.  Here's the results (using uhci.o with unlink
> > optimzations turned on)  I seem to remember a uhci driver 
> that was MUCH
> > faster at one time.  Looks like usb-uhci is a bit faster 
> and the usbdevfs
> > vs. kernel space very comparable
> > 
> > ===========================================================
> > With uhci.o (with unlink optimizations turned on)
> > 
> >            USBDEVFS                 Kernel Space Driver
> > End Time   14:28:57      End Time   14:46:34
> > Start Time 14:15:07      Start Time 14:32:32
> >            ---------                --------
> > Time used     13:50      Time used     14:02
> > 
> > ===========================================================
> > Now, with usb-uhci.o
> > 
> >            USBDEVFS                 Kernel Space Driver
> > End Time   14:57:03      End Time   14:52:12 
> > Start Time 14:54:50      Start Time 14:49:00
> >            --------                 --------
> > Time used      2:13      Time used      3:12
> 
> Hmm, userspace approach was even _faster_?
> 
> Anyways, does that mean that rio is going out of the kernel?
>                                                               Pavel
> -- 


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to