help
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 05, 2002 4:02 AM
To: [EMAIL PROTECTED]
Subject: linux-usb-devel digest, Vol 1 #1796 - 10 msgs
Send linux-usb-devel mailing list submissions to
[EMAIL PROTECTED]
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]
You can reach the person managing the list at
[EMAIL PROTECTED]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of linux-usb-devel digest..."
Today's Topics:
1. Re: Info to write a host driver on kernel 2.4.19 (David Brownell)
2. Re: new kernels (Greg KH)
3. Re: interrrupt URB's (Greg KH)
4. USB in Linux, over the net? (Erik Forsberg)
5. Re: USB in Linux, over the net? (Alan Cox)
6. Re: USB in Linux, over the net? (Alan Cox)
7. Re: USB in Linux, over the net? (Greg KH)
8. Re: USB in Linux, over the net? (Georg Acher)
9. Re: USB in Linux, over the net? (Greg KH)
10. Re: [PATCH 2.4.20-rc1] scanner.h: add/fix vendor/product ids (Henning Meier-Geinitz)
--__--__--
Message: 1
Date: Wed, 04 Dec 2002 09:07:33 -0800
From: David Brownell <[EMAIL PROTECTED]>
Subject: Re: [linux-usb-devel] Info to write a host driver on kernel 2.4.19
To: Sylvain Meunier <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Sylvain Meunier wrote:
> Hi,
>
> I'm currently updating/re-writing a host driver for SL811 circuit
> for kernel 2.4. I've several questions about expected works of
> host driver :
Any problems with the sl811hs driver in the 2.5 kernel? If that
doesn't work on 2.4 kernels, it should do so without much work;
it uses the old HCD structure.
> 4) How does work the urb->next.
Don't bother implementing this. Device drivers can implement what they
need by (re)submitting iso urbs in the completion functions, and
that approach doesn't hide the errors that could occur. ...
> PS: Is there a "How to write a host driver" or "what is exactly the
> work of host driver" :)
For now, UTSL. I like to think that the 2.5.50 ohci-hcd is small
enough that understanding it wouldn't be much of a distraction from
your real task (writing for different hardware).
Certainly look at 2.5, since there's at least the "usbtest" driver
that lets you test a lot of basic functionality required of an HCD.
(More test cases needed, of course.)
- Dave
--__--__--
Message: 2
Date: Wed, 4 Dec 2002 10:31:43 -0800
From: Greg KH <[EMAIL PROTECTED]>
To: Heinrich du Toit <[EMAIL PROTECTED]>
Cc: LINUX-USB-DEVEL mailing_list <[EMAIL PROTECTED]>
Subject: Re: [linux-usb-devel] new kernels
On Wed, Dec 04, 2002 at 12:01:27PM +0200, Heinrich du Toit wrote:
> Hello
>
> I'm writing a usb driver.. with kernel version 2.4.20 at the moment.
> I just like to know.. Will I have to make serious changes to my driver
> when kernel 2.6 is released?
Yes you will, but I don't think they would be "serious".
> Or will it keep on working?
It probably will not even compile :)
But don't wait for 2.6, try out the 2.5 tree right now. That should
give you an idea of what is happening in the USB code. What kind of
driver are you developing? Also remember that unless you have a 2.5/2.6
version of your driver, a 2.4 version will not be accepted to the kernel
tree.
thanks,
greg k-h
--__--__--
Message: 3
Date: Wed, 4 Dec 2002 10:35:17 -0800
From: Greg KH <[EMAIL PROTECTED]>
To: Heinrich du Toit <[EMAIL PROTECTED]>
Cc: LINUX-USB-DEVEL mailing_list <[EMAIL PROTECTED]>
Subject: Re: [linux-usb-devel] interrrupt URB's
On Wed, Dec 04, 2002 at 10:46:05AM +0200, Heinrich du Toit wrote:
>
> Why did they remove the statement "typedef struct urb * purb_t; " in
> kernel 2.4.20 ??
Because stupid typedefs like that are evil, and not proper kernel coding
style :)
thanks,
greg k-h
--__--__--
Message: 4
To: [EMAIL PROTECTED]
From: Erik Forsberg <[EMAIL PROTECTED]>
Date: 04 Dec 2002 18:34:38 +0100
Subject: [linux-usb-devel] USB in Linux, over the net?
Hi!
I'm in the process of choosing what to do as a Master thesis, and one
of the alternatives is to try to get USB working over a network.
The company I'm doing the work for would like to transfer USB from a
thin client (based on Linux) to a fat server (also running
Linux). That way, you could sync a Palm pilot connected to the client
using a sync software on the server and do similar stuff.
Now, my question is: Is this possible to do without sacrifying goats
every full moon for the rest of my life?
The second question is, is it possible if we keep in mind that it
shouldn't take me more than about 10-12 weeks, and that I have no or
little experience of Linux kernel programming and/or USB?
Regards,
\EF
--
Erik Forsberg http://www.lysator.liu.se/~forsberg/
GPG/PGP Key: 1024D/0BAC89D9
--__--__--
Message: 5
Subject: Re: [linux-usb-devel] USB in Linux, over the net?
From: Alan Cox <[EMAIL PROTECTED]>
To: Erik Forsberg <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Date: 04 Dec 2002 18:30:57 +0000
On Wed, 2002-12-04 at 17:34, Erik Forsberg wrote:
> The second question is, is it possible if we keep in mind that it
> shouldn't take me more than about 10-12 weeks, and that I have no or
> little experience of Linux kernel programming and/or USB?
Including building the needed hardware ? (USB is not master->master so
PC to PC needs magic boxen)
--__--__--
Message: 6
Subject: Re: [linux-usb-devel] USB in Linux, over the net?
From: Alan Cox <[EMAIL PROTECTED]>
To: Erik Forsberg <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Date: 04 Dec 2002 18:34:03 +0000
On Wed, 2002-12-04 at 17:34, Erik Forsberg wrote:
> Hi!
>
> I'm in the process of choosing what to do as a Master thesis, and one
> of the alternatives is to try to get USB working over a network.
>
> The company I'm doing the work for would like to transfer USB from a
> thin client (based on Linux) to a fat server (also running
> Linux). That way, you could sync a Palm pilot connected to the client
> using a sync software on the server and do similar stuff.
Doh I misunderstood.
You want to write a layer so that you can do USB over ethernet sort of.
Thats interesting (I'd love to see it so I can have my ipaq control
devices off the PC). Some things like interrupt support and
isosynchronous will be interesting.
Are there enough libusb user mode supported devices around so you can
hack libusb to do the work itself - so concentrating on the USB question
rather than worrying about the kernel programming too ?
--__--__--
Message: 7
Date: Wed, 4 Dec 2002 10:05:36 -0800
From: Greg KH <[EMAIL PROTECTED]>
To: Erik Forsberg <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Subject: Re: [linux-usb-devel] USB in Linux, over the net?
On Wed, Dec 04, 2002 at 06:34:38PM +0100, Erik Forsberg wrote:
>
> The second question is, is it possible if we keep in mind that it
> shouldn't take me more than about 10-12 weeks, and that I have no or
> little experience of Linux kernel programming and/or USB?
Heh, go try to read the USB spec at www.usb.org and realize why this
isn't going to be possible :)
Good luck,
greg k-h
--__--__--
Message: 8
Date: Wed, 4 Dec 2002 19:52:47 +0100
From: Georg Acher <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Subject: Re: [linux-usb-devel] USB in Linux, over the net?
On Wed, Dec 04, 2002 at 10:05:36AM -0800, Greg KH wrote:
> On Wed, Dec 04, 2002 at 06:34:38PM +0100, Erik Forsberg wrote:
> >
> > The second question is, is it possible if we keep in mind that it
> > shouldn't take me more than about 10-12 weeks, and that I have no or
> > little experience of Linux kernel programming and/or USB?
>
> Heh, go try to read the USB spec at www.usb.org and realize why this
> isn't going to be possible :)
Why? Simply use the URB-structure as the over-the-net protocol from A to B,
submit it at B and return the status and read data back to A oder the
socket. If it was a IN-interrupt or a IN-isochronous transfer, the data
block mutates to a stream. Closing the connection means "unlink urb". Even a
user space tool can do this (is ISO supported by usbfs?). It's just a matter
of the middleware ;-)
--
Georg Acher, [EMAIL PROTECTED]
http://wwwbode.in.tum.de/~acher
"Oh no, not again !" The bowl of petunias
--__--__--
Message: 9
Date: Wed, 4 Dec 2002 11:29:20 -0800
From: Greg KH <[EMAIL PROTECTED]>
To: Georg Acher <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
Subject: Re: [linux-usb-devel] USB in Linux, over the net?
On Wed, Dec 04, 2002 at 07:52:47PM +0100, Georg Acher wrote:
> On Wed, Dec 04, 2002 at 10:05:36AM -0800, Greg KH wrote:
> > On Wed, Dec 04, 2002 at 06:34:38PM +0100, Erik Forsberg wrote:
> > >
> > > The second question is, is it possible if we keep in mind that it
> > > shouldn't take me more than about 10-12 weeks, and that I have no or
> > > little experience of Linux kernel programming and/or USB?
> >
> > Heh, go try to read the USB spec at www.usb.org and realize why this
> > isn't going to be possible :)
>
> Why? Simply use the URB-structure as the over-the-net protocol from A to B,
> submit it at B and return the status and read data back to A oder the
> socket. If it was a IN-interrupt or a IN-isochronous transfer, the data
> block mutates to a stream. Closing the connection means "unlink urb". Even a
> user space tool can do this (is ISO supported by usbfs?). It's just a matter
> of the middleware ;-)
Sure, but it needs to be fast enough to do all of that. I remember
someone a while ago giving some numbers on this list showing that you
really couldn't do USB over ethernet due to timing issues.
thanks,
greg k-h
--__--__--
Message: 10
Date: Wed, 4 Dec 2002 20:40:05 +0100
From: Henning Meier-Geinitz <[EMAIL PROTECTED]>
To: linux-usb-devel <[EMAIL PROTECTED]>
Subject: Re: [linux-usb-devel] [PATCH 2.4.20-rc1] scanner.h: add/fix vendor/product ids
Hi,
On Wed, Nov 20, 2002 at 12:48:18AM -0800, Greg KH wrote:
> On Wed, Nov 13, 2002 at 05:16:09PM +0100, Henning Meier-Geinitz wrote:
> > +static inline void
> > +usb_scanner_inc_use(struct scn_usb_data *scn)
> > +{
> > + atomic_inc(&scn->users);
> > +}
> > +
> > +static inline void
> > +usb_scanner_dec_use(struct scn_usb_data *scn)
> > +{
> > + if (atomic_dec_and_test(&scn->users))
> > + usb_scanner_free(scn);
> > +}
>
> Hm, I really don't like "yet another usage counter" in a driver, as 2.5
> fixes a lot of this, but for 2.4, we don't have anything else to use...
> :(
If there is another solution, I could investigate if someone points in
the right direction.
Are there other problems that might stop the patch from beeing applied?
> Brian, what do you think about this patch?
Any response from Brian?
Bye,
Henning
--__--__--
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
End of linux-usb-devel Digest
