gmane.comp.java.javax.usb.devel

Re: javax-usb-ri-linux vs Linux kernel >= 3.5?Nousbfs!!!

Anson, Gary — 2013-05-08T18:23:38

Just where is USBDEVFS_DEVICES_NEW?

Regards,

Gary Anson
Senior Software Engineer


From: Tim Noell [mailto:tnoell< at >lexmark.com]
Sent: Wednesday, May 08, 2013 10:48 AM
To: Klaus Reimer
Cc: javax-usb-devel< at >lists.sourceforge.net
Subject: Re: [javax-usb-devel] javax-usb-ri-linux vs Linux kernel >= 3.5? No usbfs!!!

Changing USBDEVFS_DEVICES_NEW to /sys/kernel/debug/usb/devices allows javax-usb-ri-linux to work for me.  Hooray!
Also, thanks for the suggestions ...
FYI,
T.

On Wed, May 8, 2013 at 11:41 AM, Tim Noell <tnoell< at >lexmark.com<mailto:tnoell< at >lexmark.com>> wrote:
I found that in our deployment of the  3.5 kernel, we mount the debugfs under /sys/kernel/debug, which has the devices file under usb/devices.  So, I might just be able to change
#define USBDEVFS_DEVICES_NEW "/sys/kernel/debug/usb/devices"
and it may "just work".
Gonna give that a try - will report back to the list with my findings.
I don't think assuming debugfs being mounted under /sys/kernel/debug  is a good assumption for the javax-usb-ri-linu

Re: javax-usb-ri-linux vs Linux kernel >= 3.5? Nousbfs!!!

Tim Noell — 2013-05-08T18:24:54

lincoln 0 jni% git diff JavaxUsbKernel.c
diff --git a/src/jni/JavaxUsbKernel.c b/src/jni/JavaxUsbKernel.c
index 43510f7..6fd669b 100644
--- a/src/jni/JavaxUsbKernel.c
+++ b/src/jni/JavaxUsbKernel.c
< at >< at > -103,7 +103,7 < at >< at > int getInterruptType(void)
 #define USBDEVFS_SSCANF_NODE_OLD "/proc/bus/usb/%3d/%3d"

 #define USBDEVFS_PATH_NEW "/dev/bus/usb"
-#define USBDEVFS_DEVICES_NEW "/dev/bus/usb/devices"
+#define USBDEVFS_DEVICES_NEW "/sys/kernel/debug/usb/devices"
 #define USBDEVFS_SPRINTF_NODE_NEW1 "/dev/bus/usb/%3.03d/%3.03d"
 #define USBDEVFS_SPRINTF_NODE_NEW2 "/dev/bus/usb/%d/%d"
 #define USBDEVFS_SSCANF_NODE_NEW1 "/dev/bus/usb/%3d/%3d"




On Wed, May 8, 2013 at 2:23 PM, Anson, Gary <Gary.Anson< at >datalogic.com>wrote:

Re: javax-usb-ri-linux vs Linux kernel >= 3.5? Nousbfs!!!

Tim Noell — 2013-05-08T17:48:10

Changing USBDEVFS_DEVICES_NEW to /sys/kernel/debug/usb/devices allows
javax-usb-ri-linux to work for me.  Hooray!

Also, thanks for the suggestions ...
FYI,
T.



On Wed, May 8, 2013 at 11:41 AM, Tim Noell <tnoell< at >lexmark.com> wrote:

Re: javax-usb-ri-linux vs Linux kernel >= 3.5? Nousbfs!!!

Tim Noell — 2013-05-08T15:41:43

I found that in our deployment of the  3.5 kernel, we mount the debugfs
under /sys/kernel/debug, which has the devices file under usb/devices.  So,
I might just be able to change
#define USBDEVFS_DEVICES_NEW "/sys/kernel/debug/usb/devices"
and it may "just work".
Gonna give that a try - will report back to the list with my findings.

I don't think assuming debugfs being mounted under /sys/kernel/debug  is a
good assumption for the javax-usb-ri-linux project, but it may save my
bacon. (Arguably it is no worse than assuming other paths that it has
hardcoded in the past.  And then just document that debugfs needs to be
mounted there.   Long term, udev is probably the way to go ... (?) )

FWIW,
T.



On Tue, May 7, 2013 at 3:35 PM, Tim Noell <tnoell< at >lexmark.com> wrote:

Re: javax-usb-ri-linux vs Linux kernel >= 3.5? Nousbfs!!!

Tim Noell — 2013-05-07T19:35:22

Thanks for that link!  Unfortunately, they require java 6 and I am stuck on
Java 1.4 for other reasons ...
(sigh),
T.


On Tue, May 7, 2013 at 3:01 PM, Klaus Reimer <k< at >ailis.de> wrote:

Re: javax-usb-ri-linux vs Linux kernel >= 3.5? No usbfs!!!

Klaus Reimer — 2013-05-07T19:01:46

 > PS:  I have already found this:
 > http://trygvis.dyndns.org/~trygvis/wiki/Implement_javax.usb_with_libusb
 > which looks interesting, but I don't know how baked it is.
 > Still exploring options ...

If you need another option then you may want to check out usb4java 
(http://kayahr.github.io/usb4java/). It is a pretty young project 
implementing javax.usb and using libusb 1.0 as backend.

javax-usb-ri-linux vs Linux kernel >= 3.5? Nousbfs!!!

Tim Noell — 2013-05-07T18:32:12

Hi javax-usb-devel :

We have a framework that has used javax-usb(-ri(-linux))  for years.  Can't
get rid of the API.
But our kernel is being upgraded to >=3.5 and usbfs has gone away ( ;-(  )
Meanwhile, looking at javax-usb-linux-ri 1.0.3 (latest) source, it still
wants to find usbfs under either /proc/bus/usb (_OLD) or /dev/bus/usb
(_NEW).

Any plans to modify javax-usb-ri-linux to use sysfs or udev or libusb or
... ?

Curious,
T.

PS:  I have already found this:
http://trygvis.dyndns.org/~trygvis/wiki/Implement_javax.usb_with_libusb
which looks interesting, but I don't know how baked it is.
Still exploring options ...

Maven artifact

Klaus Reimer — 2013-05-01T19:39:48

Hello,

I want to publish a project in the official Maven repository which 
depends on the javax.usb API. Unfortunately the javax.usb JAR is 
currently not deployed to the Maven repo. Are there plans to do this?

(no subject)

Susheel Yadav — 2013-02-19T13:41:43

http://www.blochtech.com/components/com_content/googletrends.html

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb

(no subject)

Susheel Yadav — 2013-02-12T15:05:47

http://www.blueribbondogtraining.com/images/stories/lizgoogle.html

------------------------------------------------------------------------------
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013 
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb

Re: How to cancel pending syncSubmit Events?

Klaus Gradinger — 2013-02-05T08:07:38

i just checked my source on how i fixed it exactly and
"asyncSubmit(irp)" with an followed "irp.waitUntilCompleted()" was
what did the trick for me.

Kind regards,
Klaus

On Sun, Feb 3, 2013 at 7:27 PM, Dan Streetman <ddstreet< at >ieee.org> wrote:

------------------------------------------------------------------------------
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013 
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb

Re: How to cancel pending syncSubmit Events?

Dan Streetman — 2013-02-03T18:27:47

This is a 2+ year old thread, but I believe my response back then was
to use asyncSubmit instead of syncSubmit.

It may be a bug if abortAllSubmissions() doesn't wake up a pending
syncSubmit(), but I do not own this project anymore, Brian W does, he
is also on this mailing list so he may be able to provide more info.

On Sun, Feb 3, 2013 at 5:55 AM, Hardik Shah <hardik.shah< at >jetmobile.com> wrote:

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_jan

Re: How to cancel pending syncSubmit Events?

Hardik Shah — 2013-02-03T10:55:39

 
USB Handheld Barcode Scanner (USB HID Keyboard Device). Everything is fine, i've
opened the IN Endpoint using the UsbPipe and access it the following way: 
currentDui = new DefaultUsbIrp(buffer);  currentDui = new
DefaultUsbIrp(buffer);  currentDui.setAcceptShortPacket(true); 
usbScannerPipe.syncSubmit(currentDui);I receive the data i'm expecting and
everything works fine UNTIL i want to abort and cancel everything: 
usbScannerPipe.abortAllSubmissions();  usbScannerPipe.close();  usbScannerPipe =
null;  usbScannerInterface.release();  usbScannerInterface = null;The problem is
that syncSubmit immediately synchronizes using  "submitLock" (UsbPipeImp). After
that the whole thing just keeps blocking while waiting for data. When i call
abortAllSubmissions from another Thread, it too locks on "submitLock", waiting
for the submitSync method to release this lock. IF i scan another Barcode, the
syncSubmit method returns and everything is fine, otherwise it's a perfect
deadlock.What's the correct work-flow

Re: When will support for kernel above 2.6.31-18 happen?

Trygve Laugstøl — 2012-08-02T09:25:06

The implementation that I made ([1]) a while back depends on
libusb-1.0 and should work on all newer kernels. It is not a complete
implementation, and I know of one work on github that adds some
features. I can't remember what features I didn't implement, but it
worked fine for all of my projects.

[1]: https://github.com/trygvis/javax-usb-libusb1

When will support for kernel above 2.6.31-18happen?

Anson, Gary — 2012-08-01T21:42:05

We have customers that are using newer kernels and javax.usb 1.0.2 does not support the udev model.
Thanks!

Regards,

Gary Anson | Senior Software Engineer | Datalogic ADC
959 Terry Street, Eugene, OR 97402 USA
Toll Free: 800-695-5700 | Direct: 541-302-2039 | Mobile: 541-579-0500
www.adc.datalogic.com<http://www.adc.datalogic.com>



CONFIDENTIALITY NOTICE. This message is for the sole use of the intended recipient(s) and may contain confidential and/or privileged information. Any unauthorized review, use, disclosure, or distribution is prohibited. If you are not the intended recipient, please destroy all copies of the original message, including any and all attachments, and notify the sender immediately.

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and

Reg: Javax.usb (pipe.syncSubmit ) hangs up

Hardik Shah — 2012-07-09T05:08:24

Hello,

I am working on a code which supports integration of Usb device for an Embedded system. I am using javax.usb API. I was successfully able to send and receive data
from/to usb device.

However I came across a issue.
 I have an infinite thread which polls for the data (swiped by user). I am using pipe.syncSubmit() method. The problem comes when I want to shutdown the application and so I need
to close the polling thread, but when I closed the thread pipe.syncSubmit just hangs up (as it is waiting to read the data). I tried to release the interface  interf.release(); but it also hangs up.  Using pipe.abortAllSubmission( ) doesn't help, So I am not able to close the pipe properly.

If Polling thread or pipe is not closed properly, then restarting the application causes error in sending and receiving the data from usb device. As it uses already claimed interface. (as interface was not release previously since pipe is not closed successfully).

I can't use pipe.asyncSubmit method, as I am not sure when use

javax.usb.UsbPlatforException

Dinow — 2012-06-28T08:44:18

Could someone give me any advices or your experience to my puzzle

I used the javax(JSR80) to communicate with the USB-UPS at Linux
At first, the application is running okay (set/get) with the UPS via USB

But recently this USB-chip supplier changed their chip to the other one company
(Because the former one is stop manufacturing). Hence my application need to test or modify

And below is the message I got once I initiate the application (Compile is okay)
==> "javax.usb.UsbPlatformException : Error Submitting IRP : Timed Out"
In the meanwhile, my friend's application ran well with the new usb as the former one.
(His application don't need to change at all)
But his platform is Windows and used the Delphi

Kindly give me your experience or any advice to my doubts. 
And below are my concise code to the exception :"javax.usb.UsbPlatformException : Error Submitting IRP : Timed Out"

   //Set_Report Request=================   

     byte bmRequestType1 = (byte)0x21;       //HID   
   byte bRequest1 = (byte)0x09;

JSR80 on arm board

Ghassan Knayzeh — 2012-05-24T20:21:39

Hi All,

I am trying to use the java usb api on a arm processor. I can list all the device connected and I can open the pipes. However when I attempt to send any data on either bulk or interrupt based devices I get an error.

the following is the trace that I get:

[hotplug](3) JavaxUsbTopologyUpdater.c.build_device[117] Building device /dev/bus/usb/001/001
[hotplug](3) JavaxUsbTopologyUpdater.c.build_device[141] Device is hub with 3 ports
[hotplug](4) JavaxUsbTopologyUpdater.c.build_device[156] Device speed is 12 Mbps
[hotplug](4) JavaxUsbTopologyUpdater.c.build_config[222] Building config 1
[hotplug](4) JavaxUsbTopologyUpdater.c.build_interface[282] Building interface 0
[hotplug](4) JavaxUsbTopologyUpdater.c.build_endpoint[296] Building endpoint 0x81
[hotplug](4) JavaxUsbTopologyUpdater.c.build_device[189] Building device 2 attached to port 0
[hotplug](3) JavaxUsbTopologyUpdater.c.build_device[117] Building device /dev/bus/usb/001/002
[hotplug](3) JavaxUsbTopologyUpdater.c.build_device[141] Device is hub w

Re: Reg: Endpoints for reading the data

Dan Streetman — 2012-04-05T13:35:56

CC'ing my response to the list; I forgot to reply-all.

On Wed, Apr 4, 2012 at 8:07 AM, Dan Streetman <ddstreet< at >ieee.org> wrote:

------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second 
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
javax-usb-devel mailing list
javax-usb-devel< at >lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/javax-usb-devel

Re: Reg: Endpoints for reading the data

Dan Streetman — 2012-04-05T13:34:58

I'm no longer involved in this spec or implementation, Brian
Weischedel is the current owner, so he probably can help you more.
But I put some responses to obvious issues below.

On Thu, Apr 5, 2012 at 12:28 AM, Hardik Shah <hardik.shah< at >jetmobile.com> wrote:

You can't use SET_REPORT on an input request, you want GET_REPORT,
which is 0x01.


Typically you want to actually send data with a SET_REPORT command...?


You don't need to create a list to submit (although it shouldn't
hurt).  A list is only really needed for time-sensitive urbs like
isoc.


I think you're misunderstanding what the difference between
asyncSubmit and syncSubmit is.  When you asyncSubmit, you regain
control immediately.  When you syncSubmit, your thread is blocked
until the irp completes (successfully or with error).  Your code above
is getting the irp data immedaitely after asyncSubmit, which means
there is pretty much 0% chance that the device has transferred any
data yet.  If you want to write the code as above, use syncSubmit.
Alte

Reg: Endpoints for reading the data

Hardik Shah — 2012-04-04T07:32:05

Hi All,

I am working on a code which does a read/write operation on usb Device.

I am successfully able to read data at endpoint 1.     But I came across a situation where I need to read data from endpoint 0. How can I achieve this.

When I try to get number of endpoint for a device. I get one.
List eps = interf.getUsbEndpoints();
Eps.size()

Please tell me how to get multiple endpoints (ep0,ep1) form device.

Thanks
Shah

------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second 
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev_______________________________________________
javax-usb-devel mailing list
javax-usb-devel< at >lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/javax-usb-devel