USB Decoder does not function correctly.

[Codysseus]

I wrote a simple test case for the USB decoder that the offset of a given packet plus the length did not exceed the offset of the next packet. However, when I tried this on my own data I found that this did not hold true.
Here is a link to the modified usbcsv example with the test, and here is a link to the 7z that contains the csv in question. Be careful as the CSV is almost 2GB.

[azonenberg]

Just checking whether the fixes I mentioned on IRC solved this?

[Codysseus]

Hey Andrew! Thank you for following up with this. At the moment I do not have the bandwidth to debug, but I will get back to you on Tuesday (Dec 08 2020) with what I have tried. I hope you had a nice weekend!

Here is what I have done: I changed the speed of the PMA decoder to Low Speed. After this the Packet Decoder did not have anything in the packets vector. Inside the waveform were TYPE_ERROR symbols.

[Codysseus]

So one thing RE: my sample. Channel 0-Analog is D-, and Channel 1-Analog is D+. I was able to load a Low Speed sample into glscopeclient and it seemed to label the sample correctly. I flipped the channels and set it to Low Speed, now packets are emitted, but there is still an overlap issue. Here is a picture.

[Codysseus]

Hello Andrew,
Just wanted to update this. I tested the decoder again, and the decoder works in the glscopeclient. It must be something to do with the configuration of the headless program.
Bug issue closed!