AnsweredAssumed Answered

Chasing a kernel panic on BF527

Question asked by LeakyBitBucket on Aug 7, 2012
Latest reply on Aug 7, 2012 by LeakyBitBucket

Hardware: eval-sdp-cb1z rev 1.3

 

Software: VDK C++ project using VisualDSP++ 5.0 Update 10

 

Thread count: 2

  First thread is a main thread that currently does little more than toggle an LED on the eval-sdp-cb1z.

  Second thread is deals with USB traffic, serving as a link to a host such as a Windows PC or an ARM based board like a beagle board.

 

I have inherited some code that was "working a month ago but now it isn't" and I am looking for ideas. The kernel panic seems to be triggering on this line of code which is one of the first things the USB thread does:

 

     adi_usb_CoreInit((void*)&Result);

 

KernelPanicCode = 0x00000007 = 7

This maps to kStackCheckFailure which indicates that one of my threads is exceeding its stack.

 

KernelPanicError = 0x80000053 = 2147483731

This does not appear to be a valid value.

 

KernelPanicValue = 0x00000024 = 36

Not sure what this number signifies. Could be related to the Blackfin's RETS or maybe a thread ID.

 

Note: I do not get the error when I comment out (i.e. disable) the code that spawns the USB thread.

 

Any ideas? Should I check EBIU registers? I am still going through this code and there's a lot of things to verify such as the LDF file but could anyone give me their interpretations on the three Kernel Panic variables.

Outcomes