JVM Crash (SIGSEGV/SIGACCERR)

[Tim Cronin]

Hello,

This does not occur on Windows, but does occur on Linux.  It also occurs intermittently, but often (meaning when it happens it's not always in the same spot and it doesn't always happen).  I've seen it crash is two separate places since analyzing.  Is it a problem to pass context/devices/handles as references - I'm still closing them when necessary?  What else am I missing?

Main Error (STDERR) 

# A fatal error has been detected by the Java Runtime Environment:

#

#  SIGSEGV (0xb) at pc=0x00007f3cb30d071f, pid=5313, tid=139898936805120

#

# JRE version: Java(TM) SE Runtime Environment (8.0_72-b15) (build 1.8.0_72-b15)

# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.72-b15 mixed mode linux-amd64 compressed oops)

# Problematic frame:

# C  [libusb4java.so+0x1c71f]  libusb_submit_transfer+0x1ef

#

# Core dump written. Default location: /tmp/core or core.5313

#

# An error report file with more information is saved as:

# /tmp/hs_err_pid5313.log

In the hs_err log:

 ---------------  T H R E A D  ---------------

Current thread (0x00007f3ceb14b000):  JavaThread "printManagerExecutorService-1" [_thread_in_native, id=5505, stack(0x00007f3cc25ec000,0x00007f3cc26ed000)]

siginfo: si_signo: 11 (SIGSEGV), si_code: 2 (SEGV_ACCERR), si_addr: 0x00007f3d003b0010

[...]

Stack: [0x00007f3cc25ec000,0x00007f3cc26ed000],  sp=0x00007f3cc26e9720,  free space=1013k

Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)

C  [libusb4java.so+0x1c71f]  libusb_submit_transfer+0x1ef

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)

j  org.usb4java.LibUsb.bulkTransfer(Lorg/usb4java/DeviceHandle;BLjava/nio/ByteBuffer;Ljava/nio/IntBuffer;J)I+0

j  com.company.connections.usb.UsbConnection.sendData([B)V+69 

[...] 

System Info

USB4Java 1.2.0

java -version

java version "1.8.0_72"

Java(TM) SE Runtime Environment (build 1.8.0_72-b15)

Java HotSpot(TM) 64-Bit Server VM (build 25.72-b15, mixed mode)

cat /etc/redhat-release

Red Hat Enterprise Linux Server release 6.7 (Santiago)

Two places where this has occurred in code

public static DeviceHandle createDeviceHandle(Device device) throws LibUsbException {
    DeviceHandle handle = new DeviceHandle();
    if(device != null){
        int result = LibUsb.open(device, handle);  // <=== RIGHT HERE
        if (result != LibUsb.SUCCESS) {
            throw new LibUsbException("Unable to open USB device", result);
        }
    } else {
         LOG.error("Unable to create a USB device handle because no device was specified.");
    }
   
    return handle;
}

try {
    ByteBuffer buffer = ByteBuffer.allocateDirect(data.length);
    buffer.put(data);
    IntBuffer transfered = IntBuffer.allocate(1);
    int result = LibUsb.bulkTransfer(handle, endpoint, buffer, transfered, connTimeout); // <=== RIGHT HERE
    if (result != LibUsb.SUCCESS) {
        throw new LibUsbException("Bulk transfer failed", result);
    }
} catch (LibUsbException e) {
    String detailMsg = "Failed to send data to device: " + usbConnectionData.getDisplayName() + ". Error: "+ e.getLocalizedMessage();
    LOG.error(detailMsg);
}

[Tim Cronin]

I've now seen it in LibUsb.close() as well

Stack: [0x00007fdcdaecf000,0x00007fdcdafd0000],  sp=0x00007fdcdafcdea0,  free space=1019k

Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)

C  [libusb4java.so+0x19c8f]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)

j  org.usb4java.LibUsb.close(Lorg/usb4java/DeviceHandle;)V+0

public static void destroyDeviceHandle(Usb4JavaConnectionData connData){
    if(connData.getHandle() != null){
        LibUsb.close(connData.getHandle());

    }
    
    connData.setHandle(null);
}

[Tim Cronin]

More info:

I found that some of this is a timing issue.  The crashes that occur in open don't seem to happen if I add a Thread.sleep().  So, rather than discovering devices by opening a handle and closing it, I include the handle to the device along with metadata about the device.  With this, I can use the same handles to access the device multiple times.  This seems to get around the SIGSEGV issues occurring with an ACCERR.  However, now I get alternating seg faults in bulkTransfer:

siginfo: si_signo: 11 (SIGSEGV), si_code: 128 (SI_KERNEL), si_addr: 0x0000000000000000

siginfo: si_signo: 11 (SIGSEGV), si_code: 1 (SEGV_MAPERR), si_addr: 0x0000000000000008

Anybody have any ideas?  Maybe I'll have to try async transfers?

I'll keep at it, in any case.

Thanks,

Tim

On Monday, March 28, 2016 at 3:35:55 PM UTC-4, Tim Cronin wrote:

[Tim Cronin]

I'm now debugging with GDB while my program is running, but it's difficult without the debug symbols.  Is there a pre-built .so with debug symbols or is that something that I need to build?

Thanks,

Tim

[Stel Al]

Hello I have this code package prospa8wusb;

import org.usb4java.Device;

import org.usb4java.DeviceDescriptor;

import org.usb4java.DeviceList;

import org.usb4java.LibUsb;

import org.usb4java.LibUsbException;

public class Test {

public static void main(String[] args) {

findDevice((short)0x0BDA,(short)0x0129);

}

public static Device findDevice(short vendorId, short productId)

{

   // Read the USB device list

   DeviceList list = new DeviceList();

   int result = LibUsb.getDeviceList(null, list);

   if (result < 0) throw new LibUsbException("Unable to get device list", result);

   try

   {

       // Iterate over all devices and scan for the right one

       for (Device device: list)

       {

           DeviceDescriptor descriptor = new DeviceDescriptor();

           result = LibUsb.getDeviceDescriptor(device, descriptor);

           if (result != LibUsb.SUCCESS) throw new LibUsbException("Unable to read device descriptor", result);

           if (descriptor.idVendor() == vendorId && descriptor.idProduct() == productId) return device;

       }

   }

   finally

   {

       // Ensure the allocated device list is freed

       LibUsb.freeDeviceList(list, true);

   }

   // Device not found

   return null;

}

}

and this message when i compile the code

#

# A fatal error has been detected by the Java Runtime Environment:

#

#  EXCEPTION_ACCESS_VIOLATION (0xc0000005) at pc=0x0000000000000000, pid=4036, tid=10048

#

# JRE version: Java(TM) SE Runtime Environment (8.0_91-b14) (build 1.8.0_91-b14)

# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.91-b14 mixed mode windows-amd64 compressed oops)

# Problematic frame:

# C  0x0000000000000000

#

# Failed to write core dump. Minidumps are not enabled by default on client versions of Windows

#

# An error report file with more information is saved as:

# C:\Users\stelios\workspace\prospa8wusb\hs_err_pid4036.log

#

# If you would like to submit a bug report, please visit:

#   http://bugreport.java.com/bugreport/crash.jsp

# The crash happened outside the Java Virtual Machine in native code.

# See problematic frame for where to report the bug.