libusb-win32-devel

RE: [Libusb-win32-devel] USB_REQ_SET_ADDRESS not implemented

[Dan E. <Dan...@ne...>]

As Stephan says, you can't explicitly set the address from windows since
it's done by the USB stack, and if a user program tried to change it
independantly the framework would go wrong.
=20
You don't use the device address for transfers but the handle of the
device obtained from libusb.
=20
Dan.

________________________________

From: lib...@li...
[mailto:lib...@li...] On Behalf Of
Andrew Xiang
Sent: 05 May 2006 02:07
To: lib...@li...
Subject: [Libusb-win32-devel] USB_REQ_SET_ADDRESS not implemented


I notice that USB_REQ_SET_ADDRESS is not supported, And there is no
support for address for the bulkwrite/read?
=20
how come?
=20
thanks
Andrew
=20

RE: [Libusb-win32-devel] usb_set_configuration error

[Dan E. <Dan...@ne...>]

Andrew Xiang wrote:
> I don't see the set_configuration 0x0900 come in on the device.
> I see all the get_descriptor messages, and also, 0x880 to get the
> configuration, but NOT the 0x0900.=20
>=20
> I wonder why?

You'll have to provide more description than that for anyone to help.
Please clarify what the number are you are quoting, e.g. the whole 8
bytes of the request in order, or name the bytes.

Also, when do you see this happening? When you plug the device in, in
which case it's just windows interrogating the device, or when you run
your code.

Dan.

[Libusb-win32-devel] Problems building working libusb0.sys from win32-src-0.1.10.0

[Graeme G. <gr...@ar...>]

I've made a little more progress in understanding why I'm
having so many troubles getting libusb running.

I was under the impression that libusb0.sys needed the
DDK to be built, and since I hadn't configured this,
it wouldn't be built.

It seems that libusb0.sys is built anyway, and this is
what the operating system is choking on. If I
substitute libusb0.sys from the binary distribution,
it works OK, and I'm able to happily hack away at
libusb0.dll.

I'm using libusb-win32 version 0.1.10.0.
I've got MingW 3.2.0-rc-3 installed.
I've got the Latest XP DDK installed.
I'm on a Windows 2k box.

I don't understand how running "gmake DDK_PATH=<DDKDIR>"
(where <DDKDIR> os "D:/WINDDK" in my case) is going to
work. I can' find a reference to DDK_PATH anywhere,
and the build doesn't fail if the path is wrong.
It appears that the driver is being linked against
MingW/liblibntoskrnl.a, rather than anything from the DDK.

The libusb0.sys file that results is exactly the same length
as the one in the binary distribution, but the binaries differ
in 97 bytes (some presumably time stamp, but not all.)

Could the version of MingW I'm using be the problem ?
Any other ideas ?

thanks,

Graeme Gill.

Re: [Libusb-win32-devel] USB_REQ_SET_ADDRESS not implemented

[Stephan M. <Ste...@ia...>]

> I notice that USB_REQ_SET_ADDRESS is not supported, 

Addresses are always assigned by the host controller driver not by the
device driver, so there is no need for this feature.

> And there is no support for address for the bulkwrite/read?

What do you mean by this? What address?

> 
> how come?
> 
> thanks
> Andrew
> 

[Libusb-win32-devel] USB_REQ_SET_ADDRESS not implemented

[Andrew X. <and...@gm...>]

I notice that USB_REQ_SET_ADDRESS is not supported, And there is no =
support for address for the bulkwrite/read?

how come?

thanks
Andrew

[Libusb-win32-devel] missing USB_REQ_SET_ADDRESS

[Andrew X. <and...@gm...>]

I don't see the support for the USB_REQ_SET_ADDRESS? Would you please add 
the support in there?

thanks
Andrew

Re: [Libusb-win32-devel] usb_set_configuration error

[Andrew X. <and...@gm...>]

I don't see the set_configuration 0x0900 come in on the device.
I see all the get_descriptor messages, and also, 0x880 to get the
configuration, but NOT the 0x0900.

I wonder why?

-Andrew




----- Original Message ----- 
From: "Dan Ellis" <Dan...@ne...>
To: <lib...@li...>
Sent: Thursday, May 04, 2006 4:44 AM
Subject: RE: [Libusb-win32-devel] usb_set_configuration error


> Andrew Xiang wrote:
> > How come there is not control message going to the USB device when
> > doing usb_set_configuration???
>
> There is a control message going to the USB device, but windows doesn't
> allow it to be sent directly, since changing the configuration of a
> device is something which must be administered by the stack, so is done
> differently than normal control messsages.
>
> The DeviceIOControl call tells the driver to tell the USB stack to
> change the configuration.
>
> Dan.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job
easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
>
http://sel.as-us.falkag.net/sel?cmd_______________________________________________
> Libusb-win32-devel mailing list
> Lib...@li...
> https://lists.sourceforge.net/lists/listinfo/libusb-win32-devel
>

RE: [Libusb-win32-devel] usb_set_configuration error

[Dan E. <Dan...@ne...>]

Andrew Xiang wrote:
> How come there is not control message going to the USB device when
> doing usb_set_configuration???=20

There is a control message going to the USB device, but windows doesn't
allow it to be sent directly, since changing the configuration of a
device is something which must be administered by the stack, so is done
differently than normal control messsages.

The DeviceIOControl call tells the driver to tell the USB stack to
change the configuration.

Dan.

RE: [Libusb-win32-devel] USB unplugged

[Dan E. <Dan...@ne...>]

The call to bulk transfer will return an error, i.e. a return code less
than 0. It's up to you how you recover. The device behaviour depends
upon what chip you are using, but when it's plugged in again the OS will
execute a USB reset, so the device will see that.
=20
Whether you implement timeouts is completely up to you and your hardware
configuration. It's probably a good idea because if the device gets
stuck and stops consuming data, the host will keep retrying to send it,
and as long as the device has died completely, the host won't stop.
=20
Dan.

________________________________

From: lib...@li...
[mailto:lib...@li...] On Behalf Of
Andrew Xiang
Sent: 03 May 2006 21:42
To: lib...@li...
Subject: [Libusb-win32-devel] USB unplugged


What happens when I am doing data bulk transfer and the USB cable is
unplugged? How will the Host software know and recover? And how will the
firmware recover. I assume when the host software recover, it can send
over RESET tell the device to recover also?=20
=20
I guess it need to put some sort of timeout in the device also for the
RX/TX.
=20
=20
=20
-Andrew
=20

[Libusb-win32-devel] usb_set_configuration error

[Andrew X. <and...@gm...>]

How come there is not control message going to the USB device when doing 
usb_set_configuration???



static int usb_get_configuration(usb_dev_handle *dev)
{
  int ret;
  char config;

  ret = usb_control_msg(dev, USB_RECIP_DEVICE | USB_ENDPOINT_IN,
                        USB_REQ_GET_CONFIGURATION ,
                        0, 0, &config, 1, LIBUSB_DEFAULT_TIMEOUT);

  if(ret >= 0)
    {
      return config;
    }

  return ret;
}




int usb_set_configuration(usb_dev_handle *dev, int configuration)
{
  DWORD sent;
  libusb_request req;

  if(dev->impl_info == INVALID_HANDLE_VALUE)
    {
      usb_error("usb_set_configuration: error: device not open");
      return -EINVAL;
    }

  req.configuration.configuration = configuration;
  req.timeout = LIBUSB_DEFAULT_TIMEOUT;

  if(!DeviceIoControl(dev->impl_info, LIBUSB_IOCTL_SET_CONFIGURATION,
                      &req, sizeof(libusb_request),
                      NULL, 0, &sent, NULL))
    {
      usb_error("usb_set_configuration: could not set config %d: "
                "win error: %s", configuration, usb_win_error_to_string());
      return -usb_win_error_to_errno();
    }

  dev->config = configuration;
  dev->interface = -1;
  dev->altsetting = -1;

  return 0;
}



----- Original Message ----- 
From: "Stephan Meyer" <ste...@we...>
To: <lib...@li...>
Sent: Wednesday, May 03, 2006 3:47 PM
Subject: Re: [Libusb-win32-devel] simple example on bulkread/bulkwrite?


>
>> Does anyone have a simple example of bulkread/bulkwrite?
>>
>
> see attachment.
>
>>
>> Also, why no name show up for the device in the device manager? I ran the 
>> info maker and put in some names, but it did not show up on the device 
>> manager, I am running winxp.
>>
>>
>>
>> thanks
>>
>> Andrew
>>
>>
>
>
> _______________________________________________________________
> SMS schreiben mit WEB.DE FreeMail - einfach, schnell und
> kostenguenstig. Jetzt gleich testen! http://f.web.de/?mc=021192
>
> 

[Libusb-win32-devel] USB unplugged

[Andrew X. <and...@gm...>]

What happens when I am doing data bulk transfer and the USB cable is =
unplugged? How will the Host software know and recover? And how will the =
firmware recover. I assume when the host software recover, it can send =
over RESET tell the device to recover also?=20

I guess it need to put some sort of timeout in the device also for the =
RX/TX.



-Andrew

Re: [Libusb-win32-devel] simple example on bulkread/bulkwrite?

[Andrew X. <and...@gm...>]

Hi, Stephan,

Thanks!!

Also, why no unique name show up for the device in the device manager? I ran
the info maker and put in the manufacture names, but it did not show up on
the device manager, I am running winxp.


-Andrew

----- Original Message ----- 
From: "Stephan Meyer" <ste...@we...>
To: <lib...@li...>
Sent: Wednesday, May 03, 2006 3:47 PM
Subject: Re: [Libusb-win32-devel] simple example on bulkread/bulkwrite?


>
> > Does anyone have a simple example of bulkread/bulkwrite?
> >
>
> see attachment.
>
> >
> > Also, why no name show up for the device in the device manager? I ran
the info maker and put in some names, but it did not show up on the device
manager, I am running winxp.
> >
> >
> >
> > thanks
> >
> > Andrew
> >
> >
>
>
> _______________________________________________________________
> SMS schreiben mit WEB.DE FreeMail - einfach, schnell und
> kostenguenstig. Jetzt gleich testen! http://f.web.de/?mc=021192
>
>

Re: [Libusb-win32-devel] simple example on bulkread/bulkwrite?

[Stephan M. <ste...@we...>]

> Does anyone have a simple example of bulkread/bulkwrite?
> 

see attachment.

> 
> Also, why no name show up for the device in the device manager? I ran the info maker and put in some names, but it did not show up on the device manager, I am running winxp. 
>  
>  
> 
> thanks
> 
> Andrew
> 
>  


_______________________________________________________________
SMS schreiben mit WEB.DE FreeMail - einfach, schnell und
kostenguenstig. Jetzt gleich testen! http://f.web.de/?mc=021192

[Libusb-win32-devel] simple example on bulkread/bulkwrite?

[Andrew X. <and...@gm...>]

Does anyone have a simple example of bulkread/bulkwrite?

Also, why no name show up for the device in the device manager? I ran =
the info maker and put in some names, but it did not show up on the =
device manager, I am running winxp.=20

thanks
Andrew

[Libusb-win32-devel] Re: [Libusb-devel] Fundamental bug in libusb read() API ?

[Johannes E. <joh...@er...>]

On Tue, May 02, 2006, Graeme Gill <gr...@ar...> wrote:
> There seems to be a fundamental shortcoming in the libusb API
> definition. There is no way for a call to read() to timeout and
> report the number of bytes it read up to the timeout, because
> the return value is used for both status reporting, and
> reporting the number of bytes it reads.
> 
> By comparison, under MSWindows ReadFile() has a separate status
> and number of bytes return, and under Unix, the timeout is
> separated by the use of poll() or select() from the
> file read() function.
> 
> The libusb read() functions in practice seem to either read a number
> of bytes into the caller buffer then return the timeout error
> code to signal the timeout, leaving the caller
> no way of knowing how many bytes have been read into their
> buffer (MSWindows), or simply discard the bytes (OSX) and
> returning 0 rather than -ETIMEDOUT.

Yes, this is one of the things that is being fixed in libusb 1.0

int libusb_bulk(struct libusb_bulk_request *bulk, size_t *transferred_bytes);

> Linux is even more bizarre, it seems to not signal a timeout
> at all, but returns 8 (the packet size), and reads the packet
> received into the buffer, but subsequent calls to read return
> the same packet, over and over. (In fact to get it to return
> a timeout error at all, and stop re-reading the same packet,
> I had to apply the attached patch to linux.c).

What version of libusb are you using?

I'm pretty sure there was a bug similar to this that was resolved.

It's not exactly clear to me what your patch is doing since some parts
look completely wrong.

JE

[Libusb-win32-devel] Fundamental bug in libusb read() API ?

[Graeme G. <gr...@ar...>]

There seems to be a fundamental shortcoming in the libusb API
definition. There is no way for a call to read() to timeout and
report the number of bytes it read up to the timeout, because
the return value is used for both status reporting, and
reporting the number of bytes it reads.

By comparison, under MSWindows ReadFile() has a separate status
and number of bytes return, and under Unix, the timeout is
separated by the use of poll() or select() from the
file read() function.

The libusb read() functions in practice seem to either read a number
of bytes into the caller buffer then return the timeout error
code to signal the timeout, leaving the caller
no way of knowing how many bytes have been read into their
buffer (MSWindows), or simply discard the bytes (OSX) and
returning 0 rather than -ETIMEDOUT.

Linux is even more bizarre, it seems to not signal a timeout
at all, but returns 8 (the packet size), and reads the packet
received into the buffer, but subsequent calls to read return
the same packet, over and over. (In fact to get it to return
a timeout error at all, and stop re-reading the same packet,
I had to apply the attached patch to linux.c).

Graeme Gill.

[Libusb-win32-devel] newbie on usb-rs232 converter

[<JC...@so...>]
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Re: [Libusb-win32-devel] Re: it is possible use the libusb with an HID keyboard while the HID is under windows driver?

[Mark S. <ms...@ya...>]

I've written code to update firmware in a keyboard-like device.  Keyboards have an endpoint descriptor already for the LED byte.  You can write pattern sequence that is caught by the target as a programming command.  But I don't know enough about your particular device.
   
  

"Diego ." <di...@gm...> wrote:
    First of all, I would like to thank you for you quick reply to my enquiry.

  
The device is a special keyboard, it have only one key (the user of the PC have access to a one button. So,  he cannot sabotage the PC ). I need send to HID keyboard the active key configuration (for example, the button is an ENTER key or F1 key). And I need to send the firmware to the device for correct the possible bug in the firmware.

  

  best regards.

  



			
---------------------------------
Celebrate Earth Day everyday!  Discover 10 things you can do to help slow climate change. Yahoo! Earth Day

Re: [Libusb-win32-devel] Re: it is possible use the libusb with an HID keyboard while the HID is under windows driver?

[Robert M. <mar...@co...>]

Diego . wrote:

> The device is a special keyboard, it have only one key (the user of the 
> PC have access to a one button. So,  he cannot sabotage the PC ). I need 
> send to HID keyboard the active key configuration (for example, the 
> button is an ENTER key or F1 key). And I need to send the firmware to 
> the device for correct the possible bug in the firmware.

Why do you develop that for yourself?
Buy a KeyWarrior chip from us http://www.codemercs.com
Our firmware is stable and several versions of the KeyWarrior are 
programmable.

The IO-Warrior 40 Starter Kit is probably what you really need. It is 
complete wiht a single button to play with.
Generating keyboard events from the button read can be done in your 
application if you need that.

[Libusb-win32-devel] Re: it is possible use the libusb with an HID keyboard while the HID is under windows driver?

[Diego . <di...@gm...>]

First of all, I would like to thank you for you quick reply to my enquiry.


The device is a special keyboard, it have only one key (the user of the PC
have access to a one button. So,  he cannot sabotage the PC ). I need send
to HID keyboard the active key configuration (for example, the button is an
ENTER key or F1 key). And I need to send the firmware to the device for
correct the possible bug in the firmware.


 best regards.

Re: [Libusb-win32-devel] Re: using libusb-win32 together with libftdi

[Cory S. <co...@mo...>]

On 4/23/06, Jonathon Love - Haltech <jon...@ha...> wrote:
> Cory,
>
> Do you have libftdi and libusb-win32 packaged in the same dll?

I have not put them together in a DLL as a Windows driver because
there is no need.  I have a number of application specific exe's as
well as an application specific DLL exposing a JNI interface -- all
statically linked with libusb and libftdi.

> I'd like to put them in the same dll, but I'd rather avoid figuring out
> all the cygwin stuff. (So could I have a copy of yours?)
>
> Does mingw generate .lib files that VC++ can use?

No, it doesn't.  Though, I don't use VC++, so I don't really know how
it may / may not interoperate with object files and libraries built
with in MinGW with GCC.

Best,
Cory

Re: [Libusb-win32-devel] libusb-win32 LGPL license enquiry

[Jonathon L. - H. <jon...@ha...>]

hmm, I'm confused. Quoting 
http://en.wikipedia.org/wiki/GNU_Lesser_General_Public_License

[q]
The most commonly used method for doing so [using an lgpl library in a 
proprietary program] is to use "a suitable shared library mechanism for 
linking". Alternatively, a statically linked library is allowed if 
either source code or linkable object files are provided.
[/q]

Surely this means I have to either have it in a dll, or provide object 
files?

Also, looking at the lgpl (section 6) it seems like I have these 
obligations too.

Any thoughts?

with thanks

Jonathon

Re: [Libusb-win32-devel] it is possible use the libusb with an HID keyboard while the HID is under windows driver?

[Robert M. <mar...@co...>]

Diego . wrote:

> I have programed an HID keyboard, but I need to send some configuration 
> bytes and the firmware that will be save into a eeprom. I'm thinking to 
> include one bulk endpoint into descriptors. But, how to communicate with 
> my "custom HID device" using the libusb? When I plug my device windows 
> uses his HID driver. Could I use libusb to perform a bulk transfer when 
> my HID device is under windows driver?

Sending data to a HID keyboard is possible with the Windows HID driver 
if you use feature reports.
Why do you want to change the firmware of a keyboard?

Have a look at http://www.codemercs.com for a nice range of USB keyboard 
controllers.

[Libusb-win32-devel] it is possible use the libusb with an HID keyboard while the HID is under windows driver?

[Diego . <di...@gm...>]

Hello,

I have programed an HID keyboard, but I need to send some configuration
bytes and the firmware that will be save into a eeprom. I'm thinking to
include one bulk endpoint into descriptors. But, how to communicate with my
"custom HID device" using the libusb? When I plug my device windows uses hi=
s
HID driver. Could I use libusb to perform a bulk transfer when my HID devic=
e
is under windows driver?


Thank,

Re: [Libusb-win32-devel] libusb-win32 LGPL license enquiry

[Linus W. <tr...@df...>]

On Mon, 24 Apr 2006, Jonathon Love - Haltech wrote:

LGPL questions? See
http://www.gnu.org/copyleft/lesser.html
http://en.wikipedia.org/wiki/GNU_Lesser_General_Public_License

> In order to pass, along with our program we must provide:
>
> - object files (.lib files for VC++) for all our code

No.

> - libusb-win32 packaged in a separate dll (so our code isn't staticly linked 
> with it)

No, static linking is OK.

> - source code for libusb-win32

No, giving a reference to where it can be obtained like saying "it's on 
http://libusb-win32.sourceforge.net/ is OK, unless you changed some of 
libusb-win32 source code yourselves, in that case you have to provide a 
pointer to where the source code can be obtained.

> - prominent notices saying that sections of our program are released under 
> the LGPL

If you by "your program" refer to a bundle including portions 
of libusb-win32 then Yes.

The typical way to do it is to say that it contains the LGPL library 
libusb-win32 and tell users where it may be downloaded.

> - none of our own source code

Correct, that's how LGPL works.

Linus