Re: USB drivers

From: M. Warner Losh <imp_at_bsdimp.com>
Date: Fri, 06 Aug 2004 23:12:31 -0600 (MDT)
In message: <41124C8B.2060902_at_tellme3times.com>
            Chris <chris_at_tellme3times.com> writes:
: M. Warner Losh wrote:
: 
: >
: >You misunerstand what's actually happening.  There's no case in the
: >tree where two drivers are attached, at the same time, to the same
: >device node.  There's only one set of pointers.  However, with USB,
: >there can be reasons why multiple things can attach to the same
: >driver.  The usb code tries to do smart things for devices that have
: >multiple configurations.
: >
: >With USB and multi-function devices, here's the code that we use:
: >
: >	/* First try with device specific drivers. */
: >	probe and attach driver with a config # of -1 (usegeneric = 0)
: >		return if successful
: >	/* Next try with interface drivers. */
: >	foreach valid configuration
: >		foreach interface
: >			probe and attach driver
: >	/* Finally try the generic driver. */
: >	probe and attach driver with a config # of -1 (usegeneric = 1)
: >		return if successful
: >
: >I'm not familiar with the specific instance of ulpt and unlpt.
: >  
: >
: 
: 
: You are right I do not understand. All I know is that when the system 
: boots, at some point a test is done to see if their are USB devices. If 
: devices exist then attach drivers.

True.

: What I am trying to determine is why my multifunction printer/scanner 
: receives only one of the two  drivers. Is it because the printer does 
: not respond properly? Is it because the printer is not defined? I have 
: many questions here.

Yes.  Usb is a little complicated in this area, and there are a number
of details that are hard to get right.  It wouldn't surprise me if the
current set of drivers are less than completely optimal.

: I looked in the following and just see the code for that specific 
: device. It does not test for multifunction devices. I do not see any 
: code that follows the logic above.
: 
: src/sys/dev/usb/ulpt.c
: src/sys/dev/usb/uscanner.c
: src/sys/dev/usb/usbdevs

True.  The above code is from usb/usb_subr.c in
usbd_probe_and_attach().

: If you can point me in the right direction, or to some documents on USB 
: driver writing I will try to figure this out. Any help would be 
: appreciated. I will get this to work because I need both and the effort 
: required to switch is just not convenient.

I think that the first thing that should be done is to look at the usb
configuration.  /usr/ports/sysutils/udesc_dump is a good place to
start to investigate things.  Also, the usb mindshare book is good to
understand usb and the layers of the configuation onion.  There's a
book in the works that should help, but I'm afraid that I can't say
more than that at this time.

Warner
Received on Sat Aug 07 2004 - 03:13:15 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:05 UTC