dump-c: handle further byte_update union initializers

[tautschnig]

The changes of a8592a7 made the C front-end introduce byte updates to
ensure compliance with the behaviour of popular compilers when
initialising unions. As byte updates aren't part the C language, dump-c
must make sure such union initialisers using byte updates do not end up
in the (re-)generated output.

db4f25c was the first step in dump-c in this direction, and 3a24545
introduced further rules. The CSmith test generated with random seed
1612048908, however, demonstrated that dump-c's handling still didn't
cover all cases of byte_update expressions introduced by the C
front-end. This commit now handles those further cases.

The regression test is a minified version of that CSmith test.

• Each commit message has a non-empty body, explaining why the change was made.
• n/a Methods or procedures I have added are documented, following the guidelines provided in CODING_STANDARD.md.
• n/a The feature or user visible behaviour I have added or modified has been documented in the User Guide in doc/cprover-manual/
• Regression or unit tests are included, or existing tests cover the modified code (in this case I have detailed which ones those are in the commit message).
• n/a My commit message includes data points confirming performance improvements (if claimed).
• My PR is restricted to a single feature or bugfix.
• n/a White-space or formatting changes outside the feature-related changed lines are in commits of their own.

[NlightNFotis]

Thanks for this! We had a similar issue crop up in https://github.com/diffblue/cbmc/runs/1830186932. Is that also a related issue?

On a more general note, I'm not very familiar with CSmith. Is there any material to quickly familiarise myself with it to that I can interpret those reports better, or is it better if I get to spend more time diving deeper on it?

[codecov]

Codecov Report

Merging #5799 (cfeb536) into develop (5fd8562) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff            @@
##           develop    #5799   +/-   ##
========================================
  Coverage    69.72%   69.72%           
========================================
  Files         1242     1242           
  Lines       100907   100908    +1     
========================================
+ Hits         70361    70363    +2     
+ Misses       30546    30545    -1     

Flag	Coverage Δ
cproversmt2	43.38% <ø> (ø)
regression	66.69% <100.00%> (+<0.01%)	⬆️
unit	32.29% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/goto-instrument/dump_c.cpp	79.54% <100.00%> (+0.03%)	⬆️
src/goto-instrument/goto_program2code.cpp	68.04% <0.00%> (+0.12%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5fd8562...cfeb536. Read the comment docs.

[tautschnig]

Thanks for this! We had a similar issue crop up in https://github.com/diffblue/cbmc/runs/1830186932. Is that also a related issue?

It's the very same issue and should be fixed by this PR.

On a more general note, I'm not very familiar with CSmith. Is there any material to quickly familiarise myself with it to that I can interpret those reports better, or is it better if I get to spend more time diving deeper on it?

CSmith lives here: https://embed.cs.utah.edu/csmith/ - but our scripts/csmith.sh might be all that you'd need to look at.

[martin-cs]

I'm not sure if I see how this fixes the issue and the commit message doesn't say but it seems like the kind of change that would.

[tautschnig]

I'm not sure if I see how this fixes the issue and the commit message doesn't say but it seems like the kind of change that would.

Thank you for calling this out! I have updated the text of the commit message (and also the pull request summary) to hopefully clarify what is going on here.

[martin-cs]

Much clearer. Thank you!