Dave:
Thanks a lot: )
Ethan.
From: David Brownell <[EMAIL PROTECTED]>
To: "Ethan Du" <[EMAIL PROTECTED]>
CC: linux-usb-devel@lists.sourceforge.net
Subject: Re: [linux-usb-devel] Questions on CDC EEM
Date: Wed, 5 Apr 2006 07:18:14 -0700
On Wednesday 05 April 2006 7:01 am, Ethan Du wrote:
> > >
> > > Do you mean EEM is saving endpoints?
> >
> >No; as I said, it's less demanding because it doesn't need altsettings.
>
> I know the Data Class interface of a networking device shall have a
minimum
> of two interface settings. Do you mean this?
Did you know of some other definition of altsettings in USB? Yes.
> > > Or could you explain the relationship
> > > between altsettings and hardware.
> >
> >There's no need for such a relationship, but some controllers try
> >to move such software policies into the hardware. Examples include
> >the pxa25x and pxa27x controllers from Intel, both of which make it
> >quite painful (or impossible) to use altsettings.
> >
>
> I checked the datasheet of pxa27x, and found that it supports 7
alternate
> setting for each interface.
Exactly. The hardware has no business knowing about such software
policies,
it just complicated things, especially given errata. (If you wonder why
the
pxa27x doesn't yet have a mainstream Linux-USB driver, that's a large part
of it ...) Then look also at the pxa 25x ... and _its_ errata ... you'll
notice that altsettings are disallowed there.
> I also checked the datasheet for ISP1583, it
> doesnt move alternate setting to hardware (register). So, I think even
move
> the alternate setting policy to hardware, the number of alternate
setting is
> big enough, so it is not a big deal of saving alternate setting, is it?
So
> whats the real purpose of EEM?
You already read my opinions on that.
> And will CDC Ethernet waste time interpreting data coming from data
class
> interface, comparing to EEM?
It's got a bit less header parsing overhead than EEM, and doesn't have to
cope with the "batching" of ethernet packets. I understand that since the
Microsoft USB stacks have high per-transfer overheads, they like to design
protocols that pack lots of bytes into those transfers.
- Dave
_________________________________________________________________
Don't just search. Find. Check out the new MSN Search!
http://search.msn.com/
-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
