A descrambler turns every byte into noise from the middle of a packet on. Find the first cycle where it went wrong, and name the bug.
Hardware context: The digital blocks of a serializer/deserializer: line encoder and decoder, scrambler, CRC and error-correcting code. Keep the line balanced and busy, and detect or correct the errors a fast serial link makes.
A link trains, sends three good bytes, and then delivers only noise. The log says: ERROR cycle 6: data b5, expected 13. The sender is correct. The bug is in the descrambler.
Run the failing test to see its log and waveform. Each cycle is one received symbol. The waveform shows its kind, the byte on the wire, the descrambler output and the descrambler register lfsr.
Step 1: find the first cycle where a descrambler signal shows the wrong value, and name that signal. The error at cycle 6 is where the problem surfaced, not where it started.
Step 2: choose the cause from four possible bugs. Each one is a real bug that fails in a different way. Step 3 is optional: fix the descrambler in the editor and run the tests.
Signals in the Timeline
| Signal | Width | Role |
|---|---|---|
test_number | 8 | input of each check |
Full specification
function automatic void dscr_reset(); // link reset
function automatic [7:0] dscr_symbol(input [1:0] kind, input [7:0] rx); // one received symbol
| Symbol kind | Correct descrambler |
|---|---|
| 1 COM | Loads the seed FFFFh into the register. The byte passes unchanged. |
| 2 SKP | The register holds. The byte passes unchanged. |
| 3 Any other K code | The register takes eight shifts. The byte passes unchanged. |
| 0 Data | Each bit, bit 0 first, is XORed with register bit 15. The register shifts once after each bit. |
| One shift | Shift left by one. If bit 15 was 1 before the shift, XOR 0039h into the register. |
| Signal | In each cycle |
|---|---|
| kind, rx | The symbol received at the rising edge that starts the cycle. |
| data | The descrambler output for this symbol. |
| lfsr | The register after this symbol. |
The sender follows the same rules: it reloads on COM, holds on SKP and shifts eight times on every other symbol. After a link reset the register is 0000h until the first COM.
Cycle N is the stretch where the cycle counter shows N. A signal is wrong in cycle N if its value there differs from what a correct descrambler shows.
Diagnose the Failing Run
The test below runs the buggy design. Read its log, then find where it first goes wrong in the waveform.
Log
Waveform
Click a signal at a cycle to select it. Or focus the waveform and use the arrow keys: Left and Right move one cycle, Up and Down one signal.
Step 3 is optional: fix the code in the editor, then run the tests.
Examples
Example 1
Data with Skip Symbols- Input
COM, then 39 symbols of data and SKP
- Output
wrong_bytes = 0
The register holds on every SKP, so it stays in step with the sender and no byte is wrong.
Example 2
Data with COM Symbols- Input
40 symbols of data and COM
- Output
wrong_bytes = 0
Each COM reloads FFFFh at both ends. No byte is wrong.
Example 3
Data with Other K Codes- Input
COM, then 39 symbols of data and other K codes
- Output
wrong_bytes = 0
Other K codes are not descrambled, but the register still takes eight shifts for each.
Constraints and Contract
- Keep the two function names and their arguments.
- Symbols arrive one at a time, in order.
- The first symbol after a link reset is always COM.
Hints
Hint 1
The output was right for three bytes. What kind of symbol arrived between the last good byte and the first bad one?
Hint 2
Work out by hand what the register should hold after each symbol. A COM gives FFFFh. Compare with lfsr in the waveform.
Hint 3
The first cycle where your register and lfsr disagree is the first bad cycle. Ask which row of the table the descrambler broke on that symbol.
What the grader checks: The failing run is simulated live with the faulty descrambler. The fixture simulates the correct and each faulty descrambler on the same stream. The first bad cycle and each option's first wrong value come from those runs. The optional fix is checked on five seeded streams.
5 checks on Icarus · browser samples / server checks. Run the examples first, then check the full regression. Full checks use the server judge. Downloads contain public samples only. A passing result establishes the tested cases, not a proof for all possible inputs.
See all 3 steps and the final capstone