On 11/23/07, Chris Frey <[EMAIL PROTECTED]> wrote:
> I know the configuration exists, since the same code works when
> the usb_storage module is not loaded.
>
> I've always assumed that this conflict is a fundamental issue when
> two programs try to access the same USB device (even if one is a kern
On Fri, Nov 23, 2007 at 02:45:29PM -0500, Chris Frey wrote:
> > What's the conflict?
>
> When a Blackberry Pearl is plugged in, showing product ID 0004, and
> while the usb_storage module is loaded, I get the following error
> when trying to set the configuration:
>
> Usb::Error caught: (-16, cou
On Fri, 23 Nov 2007, Chris Frey wrote:
> On Fri, Nov 23, 2007 at 02:45:29PM -0500, Chris Frey wrote:
> > > What's the conflict?
> >
> > When a Blackberry Pearl is plugged in, showing product ID 0004, and
> > while the usb_storage module is loaded, I get the following error
> > when trying to set
Hi folks,
I've just committed some changes to CVS that check for the current USB
configuration before calling usb_set_configuration().
This allows Barry to access the database at the same time as usb_storage
is loaded.
Unfortunately this longstanding bug was due to a misunderstanding of how
I be
> Are there other configurations for the devices in question?
Nope,
USB Device 0fca:0004:
+ Main Configuration
- Mass Storage Interface
Endpoint 0
Endpoint 1
- Database Interface
Endpoints 2..10
--
On Fri, Nov 23, 2007 at 09:28:17PM +0100, Simon Ruggier wrote:
> According to LDD3 chapter 13 (http://lwn.net/Kernel/LDD3/), multiple
> configuration devices are rare, and interfaces can be claimed by
> separate drivers (i.e. I have a Logitech webcam with a mic in it, and
> an ALSA driver handles t
On Thu, Nov 22, 2007 at 09:47:04PM -0500, Alan Stern wrote:
> You are mixing up "configuration" and "interface". I think you mean
> "interface" here (configurations don't have classes but interfaces do).
Oops, yes, I had those reversed.
> > The problem is that usb_storage sees the Mass Storage
If there are no devices with more than one functionally different
configuration, then why bother even checking?
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.at
Hey all, To contribute to the research to understand WTF the object
and attribute values mean when you send an attribute request command
I've polled my Blackberry Pearl for every value (some values cause the
device to reboot) so I wouldn't recommend it.
(see attached)
poll.log3
Description: Bina
On 11/23/07, Chris Frey <[EMAIL PROTECTED]> wrote:
> I'm trying to determine if there is a way to read the current config
> through libusb, in order to avoid calling set_configuration() if it is
> already using #1.
Are there other configurations for the devices in question?
--
On Fri, Nov 23, 2007 at 09:43:49PM +0100, Simon Ruggier wrote:
> Are there other configurations for the devices in question?
Some old Blackberries (like the old 7750 I have here) seem to show
two identical configurations, with the same interfaces in them.
But more recent Blackberries seem to only
On Fri, Nov 23, 2007 at 09:55:54PM +0100, Simon Ruggier wrote:
> If there are no devices with more than one functionally different
> configuration, then why bother even checking?
It turned out that it wasn't hard to check, so I coded it up anyway.
The proper logic is now coded explicitly, for both
Hi folks,
I believe Arvind was using an 8830 and reported problems with bcharge.
I've been able to do some testing and it appears that when it comes
to the reset behaviour that most Pearl-like devices require, this
particular device behaves like a Classic and resets itself.
I've added a small sl
13 matches
Mail list logo
