feat: use pytest for qemu-system tests

[theguy147]

This PR adds the feature to use pytest (including its coverage capabilities) to the qemu-system tests. I took the scripts for the userland tests as a template and modified them until things worked as I wanted to.

I also added a few, not very meaningful but exemplary, tests for the kernel commands kbase, kchecksec, kcmdline, kconfig, kversion, slab list and slab info.

I have not yet taken a look at the necessary changes to the github workflows yet so I open this PR as a Draft for now.

EDIT: now I also implemented the changes for the github workflow.

The failing test tests/test_loads.py::test_loads_binary_with_core_without_crashing on the archlinux docker is unrelated to this PR.

[codecov-commenter]

Codecov Report

Merging #1679 (683d440) into dev (e6037f5) will increase coverage by 0.67%.
The diff coverage is n/a.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

@@            Coverage Diff             @@
##              dev    #1679      +/-   ##
==========================================
+ Coverage   59.12%   59.80%   +0.67%     
==========================================
  Files         175      175              
  Lines       21057    21057              
  Branches     1866     1866              
==========================================
+ Hits        12451    12594     +143     
+ Misses       8003     7848     -155     
- Partials      603      615      +12     

see 17 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[theguy147]

lol, the PR decreases the coverage by 56.41%? I guess this is only the coverage for the kernel commands now.

Either I messed something up, or CodeCovs promise does not hold :D

No setup required. Automatic report merging for all CI and languages into a single report.

EDIT: ah, the report was now corrected automatically...

[theguy147]

I just noticed that it might also be useful to only execute tests for certain architectures or kernel types later on, so I passed them through to the test via env vars.

Conditional tests could be executed e.g. like this:

import os
import pytest

ARCH = os.getenv("PWNDBG_ARCH")
KERNEL_TYPE = os.getenv("PWNDBG_KERNEL_TYPE")

@pytest.skipif(ARCH not in ["x86_64"], reason="")
def test_only_on_x86_64():
    raise Exception("This test is only executed on x86_64")

[disconnect3d]

Or: self.collected.extend(item.nodeid for item in items)

[disconnect3d]

This is awesome work. Thanks!