0x03. C - Debugging

[Rosicler Kleckner]

Im using a BlueNRG-2 Evaluation Board in network coprocessor mode with the GUI. Firmware is DTM v3.1. When trying to set up a secure paired connection, I sometimes get an event ACI_GAP_PAIRING_COMPLETE_EVENT with a Status code of 0x03. Status codes of 0,1, and 2 are valid. So what does code 3 mean? And the reason code is always zero.

From experimentation, one cause is an invalid authentication configuration. But another cause appears to be an uncommon situation where the remote phone is bonded but the BlueNRG-2 does not have Bonding_Mode turned on with the ACI_GAP_SET_AUTHENTICATION_REQUIREMENTS command. So the phone thinks the device has a key from a previous connection when it really doesn't.

So what is the procedure for re-pairing and exchanging LTKs after receiving a status code 3, LTK missing on local device? I've tried several things but none seem to work with an iPhone that was previously bonded.

I've tried sending an ACI_GAP_SLAVE_SECURITY_REQ to try to exchange a new key but I still get back the same ACI_GAP_PAIRING_COMPLETE_EVENT with status code 3. I tried inserting an ACI_GAP_ALLOW_REBOND before the ACI_GAP_SLAVE_SECURITY_REQ but that was flagged as "not allowed".

No. We're a month away from submitting to an EU organization for privacy rules review and I can bond with iOS but not Android. Right now, procedure is to forget the device on the phone and pair all over again. That clears the problem.

Ensure that the baud rate setting on both the transmitting and receiving sides of the UART are the same. Mismatched baud rates can cause framing errors.

Make sure that the data format settings (start/stop bits, parity, etc.) are the same on both sides.

Check that the physical connection of the UART is secure and has no interference.

If you have already attempted to re-initiate the UART but have failed, it could be due to another issue such as incorrect initialization or configuration settings. I would recommend reviewing your code to see if there are any other issues that may be causing the UART to fail. Additionally, you may want to check the error logs or debugging information to see if there are any other clues to the cause of the failure.

But the LPUART_DRV_GetReceiveStatus() just can be used to monitor the status of reception. If LPUART_RX_OVERRUN, the reception process will be interrupted. It can't recover by itself. How can I avoid this error occurs.

Hi @danielmartynek I tried using DMA that is not effective to avoid this issue. My crystal oscillator is 8MHz, the baud rate is 460800. The period of communication is 200ms. Is there any other solution to solve this issue from software? The chip is S32K144

I had someone convert a schematic I had tested/prototyped with an arduino nano and wiz5500 board, along with some IO to a custom PCB. I now finally have my custom PCB and am trying to program it, eventually aiming to connect it to the Blynk cloud. I am having trouble programming however. I tried following this guide using an Arduino Mega I have lying around.

What is the MCU: ATmega328 (as shown) or Atmega328P (as say in the Nano) ? This may be relevant to the invalid signature issue.

You appear to be trying to load a bootloader onto it but you have not broken out an FTDI header and there is no capacitor for the DTR pin of the FTDI (or similar) adaptor so it is not clear what the purpose of the bootloader would be.

I'm just trying to upload my program to the board, via my Arduino Mega Board. From looking online, this method seemed to be the way so hence I was following it. I am not too familiar with FTDI. I do just have an ICSP header on board for I/O and nothing else.

Loading a program (sketch) over the ICSP header should work. Loading the bootloader initially will set the fuses so it is correct that you try it. The FTDI header is optional but at least TX and RX would be useful so you can use the serial monitor for debugging.

However, you still have the problem that the ATmega328 which you are using has a different device signature to ATmega328p. Some Arduino cores can handle that, for example miniCore GitHub - MCUdude/MiniCore: Arduino hardware package for ATmega8, ATmega48, ATmega88, ATmega168, ATmega328 and ATmega328PB but the official Arduino core does not. But before you proceed further, can you show a bit more of the error messages that you got in post #5 including the signature that it actually found (if any).

Does my target need to be connected for this? Also - I just realized perhaps the board I was trying is blown. I have multiple so its fine, it was supposed to be designed for 12V input but I now notice no LED's are lit. I will retry with a lower rated voltage supply

Yes. The process you are following is attempting to load a program (in this case a bootloader) into the flash memory of the target ATmega328. When you select a board type in the IDE, this is naturally the ATmega328. If you select a Uno for example, which is close, and it gets further, you'll get a much more explicit error message saying what signature was found and what it expected. At that stage, you have then to load a core which can handle the ATmega328.

E.1 printf() Don't underestimate the value of printf(). The wayprintf() is implemented in Pintos, you can call it frompractically anywhere in the kernel, whether it's in a kernel thread oran interrupt handler, almost regardless of what locks are held.

printf() is useful for more than just examining data.It can also help figure out when and where something goes wrong, evenwhen the kernel crashes or panics without a useful error message. Thestrategy is to sprinkle calls to printf() with different strings(e.g. "", "", ...) throughout the pieces ofcode you suspect are failing. If you don't even see printed,then something bad happened before that point, if you see but not , then something bad happened between those twopoints, and so on. Based on what you learn, you can then insert moreprintf() calls in the new, smaller region of code you suspect.Eventually you can narrow the problem down to a single statement.See section E.6 Triple Faults, for a related technique.

E.2 ASSERT Assertions are useful because they can catch problems early, beforethey'd otherwise be noticed. Ideally, each function should begin with aset of assertions that check its arguments for validity. (Initializersfor functions' local variables are evaluated before assertions arechecked, so be careful not to assume that an argument is valid in aninitializer.) You can also sprinkle assertions throughout the body offunctions in places where you suspect things are likely to go wrong.They are especially useful for checking loop invariants.

E.4 Backtraces When the kernel panics, it prints a "backtrace," that is, a summaryof how your program got where it is, as a list of addresses inside thefunctions that were running at the time of the panic. You can alsoinsert a call to debug_backtrace(), prototyped in, to print a backtrace at any point in your code.debug_backtrace_all(), also declared in , prints backtraces of all threads.

The addresses in a backtrace are listed as raw hexadecimal numbers,which are difficult to interpret. We provide a tool calledbacktrace to translate these into function names and sourcefile line numbers.Give it the name of your kernel.o as the first argument and thehexadecimal numbers composing the backtrace (including the 0xprefixes) as the remaining arguments. It outputs the function nameand source file line numbers that correspond to each address.

If the translated form of a backtrace is garbled, or doesn't makesense (e.g. function A is listed above function B, but B doesn'tcall A), then it's a good sign that you're corrupting a kernelthread's stack, because the backtrace is extracted from the stack.Alternatively, it could be that the kernel.o you passed tobacktrace is not the same kernel that producedthe backtrace.

Sometimes backtraces can be confusing without any corruption.Compiler optimizations can cause surprising behavior. When a functionhas called another function as its final action (a tail call), thecalling function may not appear in a backtrace at all. Similarly, whenfunction A calls another function B that never returns, the compiler mayoptimize such that an unrelated function C appears in the backtraceinstead of A. Function C is simply the function that happens to be inmemory just after A. In the threads project, this is commonly seen inbacktraces for test failures; see pass() Fails, for more information.

Call stack: 0xc0106eff 0xc01102fb 0xc010dc22 0xc010cf67 0xc01023190xc010325a 0x804812c 0x8048a96 0x8048ac8.You would then invoke the backtrace utility like shown below,cutting and pasting the backtrace information into the command line.This assumes that kernel.o is in the current directory. Youwould of course enter all of the following on a single shell commandline, even though that would overflow our margins here:

The first line in the backtrace refers to debug_panic(), thefunction that implements kernel panics. Because backtraces commonlyresult from kernel panics, debug_panic() will often be the firstfunction shown in a backtrace.

The second line shows file_seek() as the function that panicked,in this case as the result of an assertion failure. In the source codetree used for this example, line 405 of filesys/file.c is theassertion

The remaining lines are for addresses below PHYS_BASE. Thismeans that they refer to addresses in the user program, not in thekernel. If you know what user program was running when the kernelpanicked, you can re-run backtrace on the user program, likeso: (typing the command on a single line, of course):

backtrace kernel.o Call stack: 0xc0106eff 0xc01102fb 0xc010dc220xc010cf67 0xc0102319 0xc010325a 0x804812c 0x8048a96 0x8048ac8. E.5 GDB You can run Pintos under the supervision of the GDB debugger.First, start Pintos with the --gdb option, e.g.pintos --gdb -- run mytest. Second, open a second terminal onthe same machine anduse pintos-gdb to invoke GDB onkernel.o:(8) pintos-gdb kernel.oand issue the following GDB command: target remote localhost:1234Now GDB is connected to the simulator over a localnetwork connection. You can now issue any normal GDBcommands. If you issue the c command, the simulated BIOS will takecontrol, load Pintos, and then Pintos will run in the usual way. Youcan pause the process at any point with Ctrl+C.

3a8082e126