On Tue, May 02, 2000, Randy Dunlap <[EMAIL PROTECTED]> wrote:
> Achim is correct -- I couldn't help him much with things
> like the CDROM drive makes noise when the usb-uhci driver is
> loaded or how USB keeps fs modules from loading.
> 
> However, he had one point which he didn't raise here, but I
> thought was the most important one for all of us.
> Maybe there's already a solution to this.  Is devfs one
> solution?  And is there a pre-devfs solution?
> 
> The problem is how to connect/match a /dev/videoN device
> to a USB camera and then how to tell your app which /dev/video
> to use.  I don't personally have more than one camera, but
> if I did, I expect that I would be having problems with this.
> 
> Think about Plug-and-Play here (but not ISA PNP).  And think
> about users who aren't developers.
> 
> This was essentially repeated as a TODO in last week's
> announcement of the [Compaq] PJB-100 device: <quote>
> 
> * USB device discovery.  There's no easy way on Linux to
>   enumerate USB devices and decide what /dev files they
>   correspond to.  We think this is a bigger issue than
>   the PJB, of course. </quote>

You missed the big flamewar^Wdiscussion about this on linux-kernel a
couple of weeks ago.

Alan Cox seemed to imply this was a userspace problem.

I didn't buy that solution for the simple reason that there are too many
different types of busses (ISA, PCI, USB, IEEE1394, etc) to require user
space applications to know about them all and know enough to probe the
device to find the distinguishing characteristics.

I argued that abstraction layers are there for a reason, to abstract
away all of that.

Unfortunately, some devices don't have any way to distinguish between
them. The best we can do on USB for devices without serial numbers is
mapping the bus topology to see which port the device is plugged into.
This is far from fool proof tho.

However, the eventual solution we came with was creating symlinks in
/dev from a userspace daemon (devfsd or something else).

JE


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

Reply via email to